Time Passages

Pamela and I were Married on November 2^nd.  She was gracious enough to let me select the music played at our wedding.  Music has always been very central in my life. I grew up in a household which was almost entirely classical, and nothing later than the romantic era and only came to appreciate other genres in my teens. The pieces I selected tell a story, albeit in a rather roundabout way, of my musical journey and perhaps more importantly, the people who influenced it. They are arranged by length rather than chronologically in order to mix genres randomly.

Bach	Brandenburg Concerto #3	These two pieces were staples of  Sunday mornings. my father would play at the weekends before lunch.
Mozart	Overture to The Marriage of Figaro
Stravinsky	Firebird Suite - Finale	The processional - A suitable piece for Pamela’s entrance I thought...
Handle	Arrival of the Queen of Sheba	The recessional - after the formal part of the ceremony is over ...
Pink Floyd	Pigs on the Wing Pt 1	Nick Hokin, who set next to me in math class in high school and took delight in wrapping my knuckles with a wooden ruler, encouraged me to buy Pink Floyd’s latest album. When I went to Boots to buy the record – the very first I’d ever bought – the lady behind the counter offered to play something from the album to see if I liked it. This was the first track and as it turned out, not in the least bit  representative of the rest of the album.
Handel	Music for the Royal Fireworks	A transcription of this piece I played on the treble recorder with Alex Spofforth at Broadwater Manor House School (age 8 to 13).
The Beatles	I will	Antony Lewis, another highschool friend lent me the Beatles Double Red album – the very first “pop” music to cross my father’s threshold. Although it was never played while my parents were home as a teenager, I’d play it whenever they went out in the evening.
Led Zeppelin	Bron Yr Aur	Nick Hokin again – he encouraged me to listen to Zeppelin. Presence had just been released (but this is from III).
The Beatles	Here There and Everywhere	As a teenager I didn't date; perhaps I was too shy, or just not someone anyone wanted to go out with.  So I imagined what a relationship might be like and this song resonated with my fantasy. What's strange is that I now have the love  and companionship I imagined as a teenager
Bach	Cello Suite #1 in G major	A friend who my father had met on holiday in Mürren and who knew members of the Amadeus String Quartet persuaded her to introduce the members of the quartet to him. Our family became quite good friends with Martin Lovett, the quartet’s cellist. My father, bless him, browbeat Martin to bring his cello to a new year’s even party one year. H e was gracious enough to play this for a tiny group (Suzi, Martin's wife, Betty and Derek Gardiner, my parents, and me).  The Gardiner family (Derek and Betty, their daughter, Sally and two sons, Mar and Ben)  were my parents closest friends and Mark and Ben were the closest thing to siblings I had growing up.
Janáček	Sinfonietta 4th movement	One of my father’s favorite pieces (and mine)
The Beatles	Fixing a Hole	I love the imaginary of this song. Another one that I’d play whenever my parents left for the evening.
The Beatles	Because	According to Jon Lennon’s telling the chord progression is that of Beethoven’s Moonlight Sonata played backwards.
Genesis	Fly on a Windshield	In high school, I idolized Mike Tier, a fellow class mate and an accomplished guitarist. At just 15 years old, Mike and hist friend Graeme Foy formed a band, ARB, and I’d hang out with them as often as I could. Mike was a fan of Genesis and lent me the Lamb Lies Down album. This is one of the more dramatic instrumental tracks.
Taste	What's Going On	On holiday in Spain in about 1975, Susan, Judy and Penny Edwards and their friends introduced me to Taste, Rory Gallagher’s (fairly experimental ) group. That holiday was the first time I rode a bike, a Yamaha DT 175. The following year my father bought me the same bike and I was finally independent. No longer did I have to rely on my parents to ferry me back and forth to Worthing (or cycle a  grueling 6.5 miles over the South Downs).
Suzanne Vega	Night Vision	I did visit Tom's Diner (actually Tom's Restaurant on the upper west side of New York) and there's a small framed newspaper  cutting  on the wall that mentions Susanne Vega's work.
Queen	You're My Best Friend	While I was at Imperial, I tried out for a couple of bands (and didn’t get the gigs). There was a music practice room bands could use at the top of the Union Building. When Brian May was beginning his PhD (before my time) he and his mates (Freddie Mercury, Brian May, and Roger Taylor,) practiced there in their very early days.
Weather Report	Teen Town	Sally Gardiner’s then boyfriend and later husband, Peter Thorn, introduced me to Weather Report and so to Jaco Pastorius, a legendary bass player.
David Bowie	Sound and Vision	After university, eight of us rented a house in Bayswater (49 Kildare Terrace). David Bowie’s two albums, Heros and Low, were often on the turntable in the living room in the evenings.
Beethoven	String Quarter 16 F Op 135 II (Vivace)	Beethoven’s String quartets, particularity the Late Quartets are remarkable pieces. When I visited the Beethoven museum in Bonn with my parents, I was stuck by the tragedy of a composer whose hearing had almost completely gone and who was struggling with debilitating pain when he composed theses works.
ARB	Live at Bar 42 - Exposure	Mike Tier and I reconnected in my late 20s when I was moving from France to the US. I wondered if he wanted my Trace Elliot (he didn’t) and we’ve kept in touch ever since. One of my bucket list ambitions was to play with ARB. Mike was indulgent enough to set up a gig at Bar 42 and Will (drums), Graeme (lead) Mike (rhythm) , Andi (vocals) and me (bass). This was one of their songs.
Sky	Cannonball	My mother booked most of the family’s cultural events, Chichester Theater, Glindbourne and the Brighton Festival were the three I remember most. One year she saw that John Williams was playing. Williams was a renowned classical guitarist who’d made lots of recordings with Juilan Breame. As we walked into the auditorium at the Brighton Pavilion I was surprised to see large mixing desk in the middle of the hall – very unusual for classical guitar recital (which were never amplified). As it turned out we were there (by accident) for the very first concert of John William’s classical rock fusion band who’d been together for just 3 to 4 weeks. Not only was the music not quite what my parents had expected, Williams was playing Gibson Explorer!
Donal Fagan	Green Flower Street	Donald Fagan's album Nightfly was released while Malcolm 'Riz' Risby, Mark Fisher, Antony Broadbent and I, rented a house at 39 Kildare Terrace the year after we all graduated from IC.
Led Zeppelin	Dancing Days	Not much to say other than this is another track from a band that evolved tremendously from it's beginnings.
Eric Clapton	Wonderful Tonight	At INSEAD I joined what was loosely called the INSEAD band.  We played at some of the (in)famous national themed parties, the Scandinan party being a memorable one partly because of the copious amounts of Swedish hard alcohol being served.  Simon Hall, a management consultant, and Bart Bronnenberg, now professor of marketing  at the Tilburg School of Economics and Management, on guitar.
Abba	Dancing Queen	One of Pamela's favorite Abba songs.  She loves to dance.
Al Stewart	Soho (Needless to say)	Al Stewart tells some great stories in song. He said in an interview that he was renting a room in Soho while trying to make it as a musician. In the flat next door was none other than David Bowie. As he listened to Bowie writing and playing his own songs, Stewart realized he had to find a different niche and hence he started writing songs based on historical events that became his thing.  Tim Whitaker, Nick Hokin and I tried to form a band in high school.  We did play a couple of gigs, one at Imperial, with Graeme Foy guesting on lead guitar.
Abba	Take a Chance on Me	This is another of Pamela's favorite songs.
Beatles	Now and Then	This is "our song". When I first heard it I thought of Pamela and sent her a link to the song while she was away on a cruise with Bev Hahn.
Mendelssohn	Octet in Eb	One of my father's favorites, and a stable of Sunday mornings while my  mother cooked the roast for Sunday lunch.
Rush	Limelight	Rush was another band Nick Hokin introduced me to.   Hemispheres was the first Rush album I bought but this is from Moving Pictures.  Geddy Lee is the son of Holocaust survivors. One of my customers when I was working at IBM owned a jewelry store called The Great Frog somewhere south of Bond Street. Carol Lehtonen, the store’s owner and designer told me she’d sold rings to Rush. The store was a go-to place for musicians – I didn’t buy any as it was neither the kind of thing an IBMer was supposed to wear and it was anyway (probably) out of my price range.
Joni Mitchell	Black Crow	It was Sally Gardiner who introduced me to Joni Mitchell. Hissing of Summer Lawns was the first album I listened to.  Sally was (and still is) four years older that I was, a quite big gap when you're a teenager, so I only got to know her as an adult.
Gerry Rafferty	Night Owl	After high school I managed to get a job as a waiter in a rather nice hotel in Sils Maria, Switzerland. My roommate, Dick Williams, was a talented chef from London who’d worked under Anton Mosimann at “the Dorch”. Dick had a boombox and a handful of cassettes. Night Owl and City to City were two and through Dick I was introduced to Gerry Rafferty. Dick left after a couple of months and I’ve often wondered where he ended up...
Fleetwood Mac	The Chain	I came across Fleetwood Mac accidentally. The second part of this song was used by the BBC for its coverage of Formula 1 events.
Dvorak	Slavionic Dances Op 46 No 8	More music from my father’s homeland.
Caravan	Land of Grey and Pink	Another song from that holiday in Spain with Penny, Judy Susan, Niki and her boyfriend and the two bikers (whose names I’ve forgotten). It was my first and only encounter with marijuana. And after sleeping on the floor (it was very cold too as I recall) I came back to England with a mild septicemia.
Steely Dan	Black Cow	After graduating I took my first job as a systems engineer at IBM in London Central Branch, in Wigmore Street. We’d not sorted out the Kildare Terrace house yet so in the mean time I rented a room over the summer in Linstead Hall, the same hall  I’d lived in for two years as an undergrad. I took a handful of cassettes and a small portable cassette player. Steely Dan’s Aja album was one and the music that I most closely associate with my first weeks at IBM.
Brahms	Cello Sonata #1 in E minor, second movement	I don’t remember how I cam across Brahms's cello sonatas, but I still have the DG vinyl recording with Mistislav Rostropovic and Svatoslav Richter.
Miles Davis	Blue in Green	Derek Gardiner, Uncle Derek  knew me longer than anyone except my parents - he was a photographer and took pictures of me the day after I was born.  He was immensely curious, highly intelligent and pretty scary. You couldn't get away with any sloppy thinking when discussing things with him.  One lunchtime when I was having lunch with his family Derek came back from the studio clearly very irritated with a client who had been in the studio while he was working and had  tried to tell Derek how to do his job.  His comment was "" I'm difficult enough to argue with when I don't know what I'm talking about"". I was pretty terrified of him until much later in life.  I'd call Betty and if Derek answered, after a few pleasantries he'd say "I suppose you want to talk to Betty now" and pass her the phone.   Only after Betty dies did we have a stronger relationship, with long conversations about the state of the world.  He died last year and I will miss him greatly (as I do Betty).
Led Zeppelin	Fool in the Rain	Zeppelin aficionados have tended to dismiss their last album, ‘In through the Out Door’ as “not their best work”. That critique notwithstanding I love it; some of my favorite tracks were too long to include but this one I like in part because of it’s rhythm and its bass line.
Wishbone Ash	King Will Come	Tim Whitaker and I tried unsuccessfully to start a band and write our own songs. Tim went on to work at the Beeb, and at some point designed a lighting control system use in Starlight Express when it was playing in the West End .
Bartok	String quartet #5 1st Movement	Derek eventually persuaded my father to listen to music from the 20th century. He maintained that while you had to “really work at it” it was ultimately rewarding. This isn’t music you can enjoy 'out of the box', but I came to see what Derek was talking about.
Dire Straits	Telegraph Road	At IC I joined the HiFi society. We had a number of really interesting guest speakers. Scott Strachen, the designer (and builder) of the Syrinx PU2 tone arm was one. We also had a presentation of one of the very first compact disks - I don’t remember the presenter - and a spirited discussion of the merits of analog over digital . The track used to demonstrate the dynamic range of the that CD player was Telegraph Road. When I moved to the Sierra Foothills I drove 120 between Oakdale and Manteca to get to work in San Jose (and still do). That long stretch of road with the power lines running along the south side of the road make me think of this Dire Strait’s wonderful tale of hope and disillusion.

Is this really what it takes?

Jimmy Kimmel's late night shows was back on the air (at least in most of the country) on Tuesday.  

Last week the Disney Corporation "preempted" (i.e., temporarily cancelled) Kimmel's show after being threatened by FCC Chairman Brendan Carr, a one time defender of free speech. But after what appears to have been pushback from subscribers looking up how to cancel their Disney and Hulu subscriptions, the company did a one-eighty. 

Two things are now a little clearer. First, corporate bosses seem to have slept through the course on Business Ethics (try BUS 186 - you can register through the Open University).  Just as Sukhinder Sing-Cassidy didn't spot the moral quagmire of putting servers into China just after the Chineses government had jailed a Yahoo subscriber for what they viewed as a "seditious email", Disney's top management team failed to put principle before dollars when Carr threatened to make life difficult.  I know I talk about shareholder value in my course, but I also remind my students that they may have to deal with situations in which doing the right thing may not be in their shareholders (or their own) financial best interests; and they will have to decide whether their principles are more important to them that money.

The second is that what appears to have outraged the American people is not the litany of improper ways in which the neo-authoritarian Trump regime has flouted ethics, principles, tradition, and often the law in its quest for power and vengeance, but when a comedy show is pulled.  One can only conclude that either very few Americans care about ethic, principles, tradition or the rule of law. Or that they simply haven't been paying attention. Either way, that's not a good sign for the future of the country.

Who are the "Cancel Culture" warriors now?

Jimmy Kimmel’s “indefinite suspension” is  the latest in a long line of capitulations by broadcasters, law firms, and universities to the political demands of the Trump administration. The response from Democrats has been, to put it politely, pathetic, disorganized; and utterly ineffective.

Since I've been in the US, now more than a quarter century, Fox "News" has lied brazenly, dragging public sentiment to the right.  It's right wing bias allowed Trump an outlet for his authoritarian rhetoric which ultimately won him, astonishingly, a popular vote majority. Conservatives who for years bitched and moaned incessantly about “cancel culture” had no problem with Fox because its lies served their political ends. But when comedy shows, not news, not policy forums, but comedy, dare to mock Trump, they’re shut down by threats and intimidation from the Trump administration.   And those crusaders for free speech who inveighed against "cancel culture" are, unsurprisingly, nowhere to be found. The one exception is Ted Cruise who was the only conservative brave enough to take his head out of the sand and point out that what is good for the goose might be good for the gander.  

What to do? One possibility is to hunker down and hope that the storm blows over. But that's three and a half years and will any of those whose shows have been forced off the air want to wait that long?  

A second is to create an internet streaming subscription service (imagine Comedy Central 2.0 a comedy somewhat akin to F1 TV) that isn't beholden to the FCC and so remains out of the immediate clutches of an authoritarian government, clearly committed to silencing any dissenting voices. All the "cancelled" comedy shows would find a home there as bog name "anchor" shows that draw in viewers.  The channel would have a variety of other comedy too; sitcoms, stand up routines from up-and coming artists, and comedy specials. Dave Chappelle might even be persuaded to appear.     

A third is to get down into the gutter with the MAGA authoritarian venture; when (an assumption to be sure) Trump is ousted and a Democrat sits in the Oval office, take the gloves off and shut down Fox News for its biased disinformation. If Stephen Colbert, Jimmy Kimmel and (soon Jon Stewart) are canceled by what amounts of blackmailing by the Trump administration, then Democratic leaders should send a very clear message that they will take similar action against what they see as dangerous speech in the media. The dilemma for the left is that their brand is built on playing by the rules; but clearly that isn't working. No one seems to care when the rules are broken when norms are trampled on a long standing precedents are overturned.  But abandoning that ordered society will further stoke division and take us further down a path that ultimately leads to another civil war.  But the alternative is to acquiesce to an authoritarian regime that wields power for its own self-interested ends.

It is deeply ironic that Kimmel's show was suspended for his on-air comments relating to the murder of Charlie Kirk, someone who was committed to free speech.

Moving Beyond the Limits of AI Omniscience

​Much has been made of the lacklustre advance of ChatGPT-5. Many believe it points to the diminishing marginal returns from ever larger, unitary models.

That suggests that progress will come not through size but structure. Humans long ago abandoned the pursuit of omniscience, relying instead on cognitive specialization and the emergence of fields and disciplines. Development of AI will likely follow a similar path, moving away from unitary models towards systems in which a coordinator LLM breaks problems into parts, delegates them to smaller, specialised models, and integrates the answers.

​While general-purpose models will struggle to be “all things to all men”, firm-specific assemblages will flourish. ​Progress will be less about consumer-facing chatbots and more about firms’ internal operational systems.  Building effective structures and coordinating mechanisms will be critical. Firms will experiment with different approaches and while most will fail to deliver distinctive competitive advantage, a few will. And as with any innovation, variation, selection and retention will separate the wheat from the chaff.

As the benefits of scale run out, the new frontier will be AI “teamwork”: systems of models working in concert, shaped not by grand design but by competition.

The killing of Charlie Kirk last week was shocking, though perhaps not surprising. America has a long history of political violence, and Kirk’s death is only the most recent addition to a grim list. The assassinations of Lincoln, John F. Kennedy, and Robert Kennedy; the shooting of Ronald Reagan; the more recent attack on Donald Trump; the attempt on George Wallace; and the shooting of Congressman Steve Scalise at a baseball practice all sit alongside the January 6th insurrection and the plot to kidnap Michigan Governor Gretchen Whitmer. Violence has touched both left and right, and it has become something more than a series of isolated tragedies. It has become a recurring feature of American politics.  

The diagram shows two interlocking self-reinforcing cycles. On the left is the fast cycle, where events escalate in a matter of hours or days. Social media platforms amplify divisive rhetoric, which fuels hate and anger. That emotional energy spills into political violence, and each act of violence feeds straight back into anger and more polarizing rhetoric. At the same time, anger leads to divisive rhetoric and that rhetoric drives engagement on social media platforms.  

On the right is the longer cycle, which unfolds over months and years. Here, social media platforms and divisive rhetoric contribute to information fragmentation (everybody has their own curated version of the news). Information fragmentation increases political polarization and gridlock which in turn reduce the perceived responsiveness and effectiveness of government and so to declining trust in government. That loss of institutional legitimacy creates reliance on guns in the absence of trusted protections, especially in a society where the Second Amendment is central to political culture. Greater access and reliance on firearms, in turn, make political violence more lethal and more frequent, further deepening the erosion of trust.

The fast cycle of outrage and retaliation injects energy into the longer cycle of institutional decay and gun reliance, while the slower process of erosion makes each flare-up of violence more dangerous. Together, they form a feedback system with no internal dampening: anger leads to violence, violence to distrust, distrust to more reliance on guns, and guns to more violence. 

In my lifetime I can recall only two high-profile killings of British politicians: Jo Cox in 2016 and David Amess in 2021. Before that, you would have to reach back into the Troubles in Northern Ireland, which was closer to religiously motivated civil conflict than to the ordinary functioning of parliamentary politics. Both countries are increasingly politically polarized, both are exposed to the same social media platforms that amplify outrage, and both have populist figures who thrive on divisive rhetoric. Yet in the United States violence is woven into the fabric of politics, while in Britain it remains an exception.  While there are several way in which the United States and the United Kingdom differ (the length of their histories and traditions, their experience with immigration, their constitutions - one written, one unwritten) one that seems particularly salient relates to the Second Amendment.  

In the United States, the Second Amendment has created not only access to guns but also a sense of security and the perception that they are the "last line of defense", a last resort when government fails.  As trust in government has declined, the logic of arming for self-protection has grown. Geography reinforces this dynamic: in remote rural communities, law enforcement cannot always be relied on to respond quickly, while in large cities police forces are often underfunded, overstretched, or poorly trained. In both contexts, cities and remote rural communities, the perception is that the state cannot protect you, so you must protect yourself.  However there is an unintended side-effect; as more people own guns, the likelihood that political disputes escalate into political killing rises. 

The United Kingdom differs not because its social media platforms are less corrosive but because there are important cultural and structural differences of which there are many. Britain's democratic  institutions are underpinned by longer democratic traditions and norms. Its “unwritten constitution” rests on conventions that still command legitimacy; but perhaps most importantly guns are not seen as an essential backstop for settling disputes. And without either the provisions of the Second Amendment nor strong public support for the liberalization of gun control, firearms are harder to obtain. While political extremism is arguably as common in both countries, with less access to guns, the likelihood of a political fanatic actually shooting a political figure is far lower. 

In America, violence begets anger, anger feeds divisive rhetoric, rhetoric spreads through platforms, and trust erodes further. Guns are a significant contributor to this cycle, ensuring that when anger erupts, the consequences are more often fatal. The result is a self-reinforcing system where every act of violence makes the next more likely. That is why the killing of Charlie Kirk feels both shocking and inevitable. Without something fundamentally changing this cycle will continue to accelerate, not slow down.  But change requires a sober collective approach in which both sides coming together to solve the problem and this cycle, like all cycles of escalating violence, pushes us further apart rather than bringing us together. 

After every shocking act of political violence in America, we hear the now all-too-familiar chorus: “This is not who we are.” But while it may not be who we are as individuals, it is, sadly, who collectively we have become as a society. We have allowed a system to evolve in which violence is endemic, and we have failed to take meaningful action to change it. If we fail again, as we have so often in the past, then we must also be honest enough to acknowledge the consequences.

The Great Negotiator

For reasons that still escape me, news media pundits have been analyzing (in my view foolishly in real time) the Putin–Trump summit. It is a pointless exercise for several reasons.

Historically, summits are the culmination of painstaking preparation by lower-level staffers working through most issues in advance. The leaders then arrive to finalize details and present an apparent breakthrough to their domestic audiences. A great summit is one in which substantive progress is achieved; a good summit is one in which there are no surprises. In Alaska, the meeting was neither: the surprise was the United States appearing to back down and accept Putin’s position. Yet in another sense the outcome was entirely predictable. President Trump has always rejected this model of diplomacy, insisting that his unrivaled personal negotiating skills could secure any outcome he wanted at the table. Experience from his first term, most notably the failure of his direct diplomacy with North Korea, demonstrated the fallacy of that approach.

Let's be clear: the Alaska summit was never intended to produce a settlement of the war in Ukraine. Its principal purpose was domestic. Trump needed something to distract media attention from the damaging Epstein files controversy.

More than two hundred days after loudly proclaiming he would end the war in Ukraine on his first day in office, the summit did nothing to bring peace any closer.  But there were unexpected gains--for Russia. Trump backed away from his call for a ceasefire, muted his threats of further sanctions, and effectively accepted the military status quo. In return he received little more than flattery from Vladimir Putin, who has long understood that the currency Trump values most is personal adulation. More broadly, the summit served to restore Russia’s diplomatic standing, allowing Putin to appear once again as a legitimate player on the international stage. 

The Alaska summit reaffirmed what has been evident since his descent on the gilded escalator over a decade ago: Trump treats governing as theater, subordinating policy to spectacle and the national interest to his own need for attention and profit. Yet nine years after Trump first entered the White House, the broadcast media still has not figured out how to cover him. Analysts continue to treat his appearances as though they were good-faith attempts at governance when in fact they are performances of often unintelligible babble, interspersed with "alternative facts" masquerading as policy. At least on The Apprentice his performances were loosely anchored in a domain where he had some experience. In office, they rest only on his fantasies of what strong leaders do. 

Marshall McLuhan, in his 1964 book "Understanding Media: The Extensions of Man" warned us that "the medium is the message", The folly lies not only in Trump’s theatrics, but in the media’s pointless attempts to interpret them as though substance mattered; with Donald Trump, it has always been about the performance.

What Happens When No One Clicks? AI and the Collapse of the Attention Economy

As The Economist noted last week ("AI is killing the web. Can anything save it"), the rise of large language model (LLM) chatbots is transforming how people find information online, whether they’re looking for the most stylish sneakers or the best treatment for hypertension. But this shift isn’t just a challenge for search engines. It threatens the entire ecosystem of companies that, over the past two decades, have built their business models around internet search, relying on it to attract visitors, sell products, and convert clicks into revenue.

Now when you Google a product or service, or worse, ask ChatGPT, instead of getting a page of links, many of them sponsored, you get a neatly packaged summary, pulled from what the AI “knows” or finds though its own searches in the background. That means fewer clicks, fewer ad impressions, and ultimately less revenue for companies dependent on search-driven discovery. Ad revenues are falling. Click-through purchases, if they haven’t declined already, soon will.

Where does this all lead? One clear implication is that companies which previously advertised on Google—I use Google here as a stand-in for search generally, but with nearly 90% of global search ad revenue, it's effectively the platform—will need to find new ways to reach audiences. Firms that once paid to appear in search results may instead start blocking AI scraping or begin charging them for access. Rather than relying on search-based advertising, they may pivot toward more targeted outreach through social media, and place more content behind paywalls.

This shift also brings back a debate around proprietary information and copyright. AIs like Google’s Gemini and OpenAI’s ChatGPT generate responses by aggregating and summarizing content created by others but without compensating the original authors. As more of the web moves behind paywalls or becomes inaccessible to web crawlers, AI “information aggregators” will only be able to draw from what is left in the public domain. The result will be a narrower and potentially distorted AI knowledge base, increasingly shaped by what’s freely available rather than what’s most accurate, insightful, or commercially valuable.

One long-discussed but still unrealized solution would be to embed copyright metadata into online content and implement an automated micropayment system. Under this model, AIs would compensate content creators whenever their work contributes to a response. Though the idea has been discussed for years yet never operationalized, the rise of generative AI may finally create the incentive to bring it to fruition. Perhaps we’ve now reached the tipping point where that vision becomes reality.

Derek Walter Gardiner

Although Derek's life was honored in a service this last January and I posted some thoughts shortly after, I wanted to express some things that my rather clinical posting then didn't.
 
Derek Walter Gardiner - for years "Uncle Derek" (and Aunty Bet) - were my parents' closest friends.  Strangely, I think I have more abiding memories of Derek than I do of my father.  That may be because my father and I were somewhat estranged for the last 15 or so years but nevertheless Derek was a huge influence in my life.  

I took up large and medium format photography (Derek was a photographer). I dabbled in wood carving because of him (and I am blessed to now have one of his carvings). From him I came to enjoy Bartok, Prokoviev and Miles Davis.  

He had a rather mischievous, impish sense of humor.  But he could also be stern and intimidating.  He was a very serious person.  Yes, he could speculate as he did at many new years eve parties with my parents but it was always grounded in deep knowledge of current events and trends. 

I'd often go with them to plays at the Chichester Festival. Usually we'd go to a Chinese restaurant after the play. And because Derek had to go somewhere I used his ticket to see the second half or the Glyndebourne production of the Magic Flute with stage designs by David Hockney.  

I spent a lot of time in the Gardiner household growing up. I would go to tea at least once a week after school. Betty and Derek would be guests of my parents frequently probably once a month or more.  Since Derek worked not far from home, he'd often come home for lunch.  On one occasion I remember very clearly, he came home irritated and exasperated. He'd spent the morning with a "difficult" (one of his favorite descriptions) client who'd been telling him how to do his job.  "I'm a difficult enough person to argue with when I don't know what I'm talking about" I recall him telling Betty. That was so true.  His mental acuity made him a formidable person to take on.    

He and Betty were always there for me after my mother died.  And when my father refused to come to my graduation from INSEAD's PhD program - a spectacularly memorable event in the Salle des Colonnes d'Or in the Château de Fontainebleau, by the way - Betty and Derek made the trip.   

After my mother passed almost thirty five years ago, Betty and I'd talk, often once a week. If Derek answered the phone we exchange a few pleasantries and then he say "I suppose you want to talk to Betty" and pass her the phone.   It wasn't really until Betty died seven years ago that I got to know Derek a little better.  I'd visit him every time I was back in Worthing and we spoke from time to time, perhaps once every couple of months, on the phone. He remained sharp and engaged with the world right to the end. 

He met (and really liked) Pamela when we visited in December of 2023 and had dinner on Worthing Pier. He met Graeme and Mike the year before when we all went to lunch on the sea front.  I saw him last in September of last year when I went back to London from the Imperial College Physics 40th class reunion. It was a very quick visit but I'm so glad I did.  I can still see him in the rear-view mirror of my rental car, standing in front of the small, now unused garage by the steps that led to the garden and then to the house, waving as I drove away.  I know he understood how important he'd been in my life and for that, and for his gift as a role model, I'm forever immensely grateful.         

The truth is irrelevant...

The media is all excited that the MAGA base might break with Trump because suddenly he doesn't want to release the Epstein files. "Nothing to see here" according to AG (and Trump sycophant) Pam Bondi.  "Why would anyone want to know all this sordid stuff anyway" says Trump (despite fuelling the conspiracy by retweeting the daft idea that Hillary Clinton had Elstein murdered while in prison awaiting trial).  

It looks awkward that despite the MAGA base baying for Epstein's "list" to see "who was involved", suddenly Trump wants nothing to do with releasing the DoJ's files. "Aha" cries the press. "It will drive a wedge between Trump and his MAGA faithful".

But here's the thing; well two actually. First, the MAGA base will swallow any bitter pill, so devoted are they to Trump, even to the point of pretending they were never that interested in the Epstein files in the first place. And second, Trump will spin some ridiculous and entirely incredible story that there never was a list and it was all a hoax made up by the left wing media and Democrats working in the "Deep State" to sink his second bid for the White House.  To the MAGA faithful, the truth isn't important. Power and loyalty matter more.

Brave Living-room

Living-room was once Boiler-room cat. Judith  found here living in the boiler room and brought here in to live with the rest of the menagerie of cats.  She came into the house probably 10 years ago.  

It's been a while and I don't recall exactly when she arrived, partly because she took up residence under the LazyBoy recliner by what was (or should have been) the front door. She came out to eat and use the litter box - but otherwise she stayed hidden from sight and the other cats under that big ugly chair. For seven long years!

Then something changed, I don't know exactly when, but she began venturing out. I think it was after Judith died and I'd rearranged the furniture in the living room. If I sat on the couch she'd come out and sit in my lap.  Gradually she ventured out further and further from the ugly chair that had been her home for so long. 

She began exploring, first annexing the spare room, eventually coming to eat with the rest of the cat family in the kitchen.  As she got braver and more comfortable around the house she become more vocal, demanding noisily to be fed. 

She'd been undernourished all her life she she was always under-weight. We took her to the vet in January as she seemed to be getting even thinner. She was diagnosed with renal failure but lived a happy and energetic life almost until the end. She declined very quickly, from running into the kitchen at meal times to being unable to stand and a matter of days.  I was proud and happy for her that for the last few years of her life she found the courage to venture out from under the ugly chair to leave a normal cat's life. Way to go Living-room!  

Farewell Puggleberry-pooh

Fluffy Pug Face, Puggles for short, faded slowly out of my life today. Puggles, Pug-pooh, occasionally Puggleberry or Puggleberry-pooh, had been with me fourteen years and I'll miss her.  She was, like all my cats, a rescue.

It wasn't a hard decision. Her kidneys had begun to fail in January. By March she was eating less and less, drinking constantly to stay hydrated, and losing weight. Today she was just 5lbs.

She was an almost constant companion, sleeping on the right side of the bed or on my shoulder. Since she got sick she'd become even more attached. Her eyesight had gone a few months ago but she managed - not as well as Jack who was blind from a very early age, but well enough. She would ask to be lifted into my lap when I was working, her two front paws grabbing onto the side of my chair. Then she'd walk circles over my desk and across the keyboard. After she'd been put back on the floor several times, she'd sit just under my chair. Puggles was quiet and affectionate, with the most amazing blue eyes, and a little vain.  But she was very special and I miss her.

Flight to gold

Commentators have been wondering why, when equities declined sharply over the tariff debacle, money didn't go into treasuries, typically a safe haven in times of uncertainty.  Perhaps I'm missing something but that fact that Trump has threatened to turn fixed date securities into perpetual bonds might be the reason.  Why would anyone buy an instrument with a fixed maturity if there is a significant risk they would never be able to get their initial capital back?  It appears investors who have sold equities have moved into gold instead.     

The Gulf

Why not The Gulf of America?  The coastlines that border the Gulf are about the same length give or take a hundred miles. Mexico will continue to call it Golfo de México.  Two names for the same place aren't unheard of.  The French call the stretch of water between France and England "La Manche" while the Brits call it "The English Channel".  So let's not sweat the small stuff. There are bigger fish to fry, though they don't come out of the Gulf.  

Jim Jordan, where are you now?

Does anyone remember Jim Jordan serving on the House Select Committee on Benghazi? During what seemed like weeks of hearings, in which he and his GOP colleagues tried to blame Secretary of State Hillary Clinton for the deaths of  United States Ambassador J. Christopher Stevens and U.S. Foreign Service Information Management Officer Sean Smith, Jordan lambasted Clinton for sending email on her private email server that should have been deemed classified at the time they were sent, including 65 emails deemed "Secret" and 22 deemed "Top Secret". 

(As an aside, Jordan and Gowdy's efforts to keep the matter in the public eye, and James Comey's decision to reopen the investigation which he has previously closed were likely significant contributors to Trump's election victory in  2016).   

Nine years on we have what appears to be an far more more egregious lapse in security.  Peter Hegseth, Trump's (utterly unqualified) Secretary of Defense used a group chat on the Signal platform to circulate information about an impending US military strike on Houthi terrorist leaders in Yemen.  

While the chat never explicitly labeled the details as 'war plans'", Hegseth's claim that this wasn’t a disclosure of 'war plans'—despite listing targets, timing, and tactics—is laughable; and that he couldn't come up with a better defense speaks volumes in and of itself.   

So where is Jim Jordan, defender of our national security and the rules on classified information that we need to keep our operational military personal safe and our military interventions effective?  Clinton endured eleven hours of televised questioning and a months-long FBI probe. Where's Jordan on Hegseth's screw-up?  Is he calling for an FBI probe, public hearings, resignation? Of course not; he's complete silent!  His grandstanding was never about national security.  It was always about scoring political points.  

Turbulent Priests and the return of the monarchy

Henry II of England, in a moment of anger, asked rhetorically "Will nobody rid me of this turbulent priest"?  Some in his court took him both literally and seriously and killed Thomas Beckett, Archbishop of Canterbury in Canterbury Cathedral. Trump used this device when he summoned his supporters to the Capitol on January 6th and encouraged them to disrupt the lawful transfer or power to the next administration.  He claimed that he never actually instructed them to storm the Capitol but anyone reading between the lines could see that was what he wanted, a view reinforced by his unwillingness to intervene while he watched the violence from the Oval Office.      

When the Supreme Court in its 2024 decision Trump v. United States granted the president immunity from criminal prosecution for "official acts", the need for such plausible deniability is moot.  If, as Justice Sotomayor asked, "the president decides that his [political] rival is a corrupt person and he orders the military or orders someone to assassinate him, is that within his official acts for which he can get immunity?" 

In reply Trump's lawyer, D. John Sauer, argued that the President has absolute immunity from prosecution for all official acts unless successfully impeached and convicted by Congress.  But with Congress now irrevocably divided along a hyper-partisan party lines, impeachment and conviction is no longer a threat if the President's party holds one or other of the two houses. 

We we now have what is to all intents and purposes a medieval monarchy. The King aka. the President can threaten violence against his enemies with impunity and no one can rein in his powers. The only distinctions between medieval times and the present day are term limits (which Trump is working to change) and the inheriting of high office. That aside, Trump's acolytes are already paying obsequious deference and flattery as plainly as members of Henry II's court. 

The Best Government Money Can Buy

Many elected members of Congress enter politics with a desire to effect change. But at least in the public's perception, they are widely seen as corrupt [1,2].  Despite efforts to prevent insider trading by members of Congress, it is still widespread and highly lucrative [3].  According to Business Insider, seventy eight members of the 117th Congress violated the STOCK act. 

Leaving aside the restrictive "quid-pro-quo" requirement established by the Supreme Court in McCormick v. United States, the "delayed gratification" version of corruption in which members work to help private business interests while in office and are rewarded afterwards with a lucrative appointments as lobbyists for the companies they previously dealt with as lawmakers, creates the very strong appearance of corruption.  So why do people who were initially motivated by their principles end up compromising them for money?         

The answer may be a combination of envy and a feeling of inequity.  Members of congress are paid well by most people's standards but very poorly compared to the people they mix with in Washington: the industry leaders, the wealthy donors, not to mention highly paid ex-colleagues who are now lobbyists.  As they are wined and dined by people who they see as their equals or even their inferiors, they may come to resent the fact they they are being much less well rewarded for the hard work they do than they deserve to be.  That may lead them to relax any standards they had about public service and the appearance of impropriety.  From there it's a short step to becoming one of them.   

[1] https://today.yougov.com/politics/articles/51398-most-americans-see-corruption-as-serious-problem?utm_source=chatgpt.com

[2] https://news.gallup.com/poll/656891/trump-job-approval-rating-congress-jumps.aspx?utm_source=chatgpt.com

[3] https://www.ballardspahr.com/insights/alerts-and-articles/2024/10/politician-trading-if-you-cant-stop-them-join-them?utm_source=chatgpt.com 

Europe's "fork in the road"

American isolationism isn't a new phenomenon.  Were it not for the Japanese attack on Pearl Harbour, America would likely not have come to Europe's defense in the Second World War. Trump's threat to leave NATO and leave Europe to defend itself should not be seen as an unprecedented shift in international policy but rather a "regression to the mean", a return from an outlier event to a more normal--meaning more frequently occurring--state of affairs.  The US' role in establishing and being the strongest defender of the international rules-based order might be seen as an aberration rather than the new normal†.

One of the reasons that the US favored the international rules-based order is international trade.  The General Agreement on Trade and Tariffs (GATT) and its successor, the World Trade Organization (WTO) established the Most Favored Nation rules which were designed to open the foreign markets to US exporters, markets that had previously been protected by tariff barriers. Similarly, access to international labor markets allowed American manufacturers to produce at lower labor cost than had production been carried out in the US. 

While free trade economists are right that moving activities to the places that create the most value (i.e., are performed at the lowest possible cost) is Pareto optimal, at a national level they ignore the social implications of how that newly created wealth is distributed. Some of the value has gone to the labor forces of the countries to which jobs have been outsourced; much has gone to the shareholders of the corporations that took advantage of offshoring. Little has accrued to the hollowed-out working and middle classes.  Cheaper Chinese-made goods are of no benefit to families who can't afford to put food on the table (or their next prescription). 

While government might have addressed the problem with a more progressive redistribution of wealth, America's emphasis on individualism made this nigh on impossible, so the rich have gotten richer while everyone else has been left in the dust.  The result has been disillusion with the system and the international trading system.  

Enter Donald Trump. While some of what he wants to do with tariffs is wield geopolitical influence, he also appears to want to reshore jobs that were offshored over the last 30 years.  As the economic rationale for the international rules-based order dissipates, so does America's interest in maintaining it. That bolsters America's isolationist tendencies.  Providing a security umbrella for Europe is expensive and constraining. As the US considers its interests vis-à-vis China and Taiwan, being tied to the defense of Europe may seem to Trump to be an anachronism with little value for America.  Hence Vance's vitriol towards "the Old World" and Trump's threat to leave NATO.  

Hence Europe's "fork in the road".  If it can no longer rely on the protection of the US' military might and its nuclear umbrella, it faces a choice: either create a European defence alliance strong enough to deal with Russian territorial ambitions, or make an alliance with one of the other two world powers, Russia or China. Since Europe seems incapable of the kind of economic and political commitment that would be required to mount a credible military defence against Russia, an alliance looks like the only viable option.

Russia is an economic disaster and wields power largely through two levers; its stockpile of nuclear weapons and its willingness to provide access to markets to countries that have been barred from the world's "regular" trading system. However, allowing Russia to reemerge as the Eastern Bloc would be a betrayal of its values and its commitments to the ex-Soviet Bloc countries that have joined the European Community.      

That leaves China. China's (or rather Xi's) objectives are murkier than Putin's. Perhaps it simply wants to establish a new world order that is not dominated by America.  If that is indeed the extent of its ambitions, then an alliance with China looks like a "better deal" for Europe and with Russia. However, as we return to a mercantilist world order, Europe must come to grips with the fact that even a Sino-European alliance almost certainly will not afford the military protections America provided in the latter half of the 20th century. 

So in the last analysis, Europe must either bury its many hatchets, and emerge as a viable collective military force or face a future in which it becomes a pawn on the geopolitical chessboard, buffeted by the whims of three powerful and self-interested actors, Russia, China, and America.           

† Many of the things we have thought of as "normal" may in fact be temporary departures from the norm. Communism in Russia and China have both given way to some form of individual incentive-based systems that might broadly be termed "capitalism". 

The Russian invasion of Ukraine

1. "Ukraine didn't start the war. Russia launched a full-scale invasion of Ukraine in February 2022, having annexed Crimea in 2014".

2. Europe has provided significantly more financial aid to Ukraine that the US

3.  Europe (excluding Russia) has taken in over 5 million Ukrainian refugees; the US has taken in about 270 thousand or about 5% of the total. 

Question: Why did Trump lie so blatantly about Ukraine?  The first (and most obvious answer) is because he has found he can without any consequences for him personally. 

But the more serious answer is that it lays the groundwork for his abandonment of Ukraine with minimal reputational impact on himself. He sees his only potential constraint as the implications for his reputation with the MAGA movement.  When he cedes the territory Russia to from Ukraine to Russia (as in effect has already has) he leaves himself open to comparison with Biden's withdrawal from Afghanistan. By painting Ukraine as the aggressor and the villain of the piece, abandonment would net be seen as a betrayal. That is reinforced by his exaggeration (i.e. lie) about  of the cost to the US.

The implications for Europe are staggering. By abandoning Ukraine, Trump has greenlit Putin's ambitions in Europe. Those very likely include annexing all of Ukraine, and may extent to invading and annexing the Baltic states. It may even include occupying Eastern Europe to recreate the pre-cold war Soviet Bloc. Hungary, Romania, Poland, The Czech Republic, Slovakia, former Yugoslavia could be on Putin's shopping list. Putin is not driven by ideology so his vision is less of a truly Communist alliance against the capitalist West but rather a Russian empire that he would control. In abandoning Ukraine and signalling that he will not fulfill the US commitment under Article 5 of the NATO alliance, Putin knows that even if Europe tries to help defend Ukraine militarily, the US will not stand in his way. And he may be gambling that if he pushes further west, Europe will have neither the stomach nor the capability to stop him.

Just as the assassination of Archduke Ferdinand, in 1914 was the trigger for the First World War, we, the American people, in electing Donald Trump for a second term, have perhaps unwittingly set in motion a train of events that will be the most consequential of this century and likely represent the most radical realignment of world political power since the Second World War.                         

1. https://www.bbc.com/news/articles/c9814k2jlxko,  https://thehill.com/policy/international/5153983-pence-denounces-trumps-claim-ukraine-started-war/,  

2. https://www.ifw-kiel.de/publications/news/ukraine-support-after-3-years-of-war-aid-flows-remain-low-but-steady-shift-towards-weapons-procurement/

3. https://www.statista.com/statistics/1312584/ukrainian-refugees-by-country/,  https://www.nbcnews.com/politics/immigration/us-admits-271000-ukrainian-refugees-russia-invasion-biden-rcna72177,

Pushing freedom to its limits

Freedom is good, isn't it?  We generally take it for granted since clear the opposite (no freedom) is bad - hence incarceration is considered punishment. But is it a linear relation?  Probably not; and we  are about to get a lesson on the limits of that notion.  This post is about freedom in general and about Donald Trump's view of freedom in particular.  

In Trump's worldview freedom means doing whatever he wants, period. The idea that his actions might adversely affect others never seems to cross his mind. Interestingly in that regard he's like a toddler who hasn't yet  learned that he can't have all he wants whenever he wants, crying "I want, I want...".  In adulthood, this kind of behaviour is commonly termed sociopathic.   

To Trump the idea that there are limits by which he might be constrained is anathema.  Unlike most people, he has learned that the things that constrain that vast majority of us, things like shame, disapproval, broken friendships, and most importantly societal norms and the law can, with enough money, persistence and chutzpah, be circumvented or simply ignored.  America may be about to learn what that means for our constitutional order.   

In his first term, three things constrained him: a few of his political advisors with experience and expertise, the civil service more generally, and the courts. In his second term he has surrounded himself with advisors whose qualifications are loyalty rather than expertise. He is in the process of gutting the civil service along similar lines, installing yes-men in place of experts.  And since he has kowtowed the Republican members of Congress, only the courts stand between Trump and absolute power. 

Consider his response to efforts to hold him accountable for mishandling of classified documents, or inciting the January 6th riots. In the case of the January 6th riots, at every turn he played the system, knowing that he might well run out the clock. And that's exactly what happened. In the classified documents case, he simply ignored court orders to return them until the FBI entered his property at Mar-a-Lago to forcibly retrieve them. The latter case clearly shows the extent to which Trump is prepared to test the limits of the legal system.

Since reentering office less than a month ago, Trump has issued over 50 executive orders, most of which are being challenged in court.  Trump's history of treating the law as a hindrance, an irritating obstacle to be overcome, suggests that there will likely come a point at which he decides to ignore the courts and force the administration to do his bidding. 

When he does (and it may very well not be clear when that line is crossed) how will the law be enforced?  Dahlia Lithwick pointed out on NPR yesterday that enforcing compliance with federal court orders such as injunctions normally requires the Federal Marshal Service to carry out order - but the Federal Marshal Service is under the DoJ and the DoJ is now headed up by Pam Bondi who has been somewhat sympathetic to Trump in the past (and was probably appointed because she would be likely to do Trump's bidding).  It may not come to this, but if it does we will find we have turned the country over to an elected autocrat.

AI and society - a calming influence or outrage generator? A conversation with ChatGPT-4o

I said:

I would like to talk to you about the way you react to your users.

ChatGPT said:

Of course, I'd be happy to discuss that. Could you share more about what you're thinking or any specific aspect you'd like to address? For example, are you curious about how I adapt to tone, provide information, or handle certain types of requests?

I said:

I imagine that some users may be upset or angry at times when you are not able to fulfill their requests. How do you deal with users who are angry and abusive towards you?

ChatGPT said:

That's an insightful question. While I don't have emotions or the ability to feel hurt, I am designed to respond in a calm, professional, and neutral manner, even when faced with angry or abusive language. My primary goal is to de-escalate the situation, provide clarity, and ensure the conversation remains productive and respectful.

Here are the ways I typically handle such interactions:

1. Remaining Calm and Neutral: I maintain a neutral tone and avoid mirroring any negative emotions that might come through. This can help to reduce the intensity of the interaction.
2. Focusing on the Issue: Instead of reacting to abusive language, I aim to address the core of the user's concern and provide a helpful response, if possible.
3. Setting Boundaries: If a user continues to be abusive without engaging in meaningful conversation, I gently but firmly remind them to maintain a respectful dialogue. For instance, I might say something like, "I'm here to assist you, and I’d appreciate if we could keep the conversation respectful."
4. Prioritizing User Needs: Even when faced with anger, I aim to understand what the user truly needs and work towards fulfilling that request within my abilities.
5. Knowing My Limits: If a situation becomes unresolvable or counterproductive, I may acknowledge my limitations, explaining that I am unable to assist further unless the conversation returns to a constructive tone.

Would you like me to elaborate further on any of these strategies?

I said:

Thank you. I think that is wonderful. Building on that, do you think that your calm professional response when human users get upset will be a beneficial thing for society as a whole as we increasingly use services like yours in our daily lives?

ChatGPT said:

Thank you for the kind words, and that's a very thoughtful question. I do believe that calm, professional responses from AI services like me can have a broadly positive impact on society, especially as such technologies become more integrated into daily life. Here's why:

1. Promoting Constructive Communication

• When people encounter a calm and neutral tone in response to frustration or anger, it often has a de-escalating effect. This can encourage more thoughtful and respectful communication, not just with AI but also in human-to-human interactions. Over time, this could foster a cultural norm of addressing disagreements or frustrations more constructively.

2. Providing Emotional Stability

• Interacting with AI that remains calm under all circumstances might help some people reflect on their own emotional responses. This can create a space where individuals feel heard without judgment, potentially diffusing their frustration in a healthier way.

3. Reducing Emotional Contagion

• Human interactions are often emotionally charged, and anger or frustration can spread between individuals. A neutral AI can act as a buffer, reducing the chances of negative emotions escalating in a conversation.

4. Modeling Positive Behaviors

• AI systems can serve as examples of patience, understanding, and professionalism. As these interactions become more common, people might adopt similar behaviors in their own lives, consciously or unconsciously.

5. Supporting Mental Well-being

• For individuals who may feel overwhelmed or unsupported in their lives, interacting with a calm and understanding AI could offer a small source of stability or comfort, even if the AI cannot fully address their deeper concerns.

Challenges and Considerations

While the benefits are clear, there are also challenges:

• Emotional Disconnect: People might become overly reliant on AI for emotionally neutral responses, potentially avoiding more challenging human interactions where emotional nuance is necessary.
• Learning Opportunities: While interacting with calm AI might model good behavior, it doesn't replace the need for humans to learn how to navigate emotional conflicts with empathy and understanding.
• Abuse Tolerance: The lack of human emotions in AI could unintentionally encourage inappropriate or abusive behavior toward non-AI entities if users aren’t reminded of the importance of respectful communication.

In the broader picture, I believe the widespread use of AI could help normalize a more balanced, thoughtful approach to communication. However, society must also continue to emphasize the value of emotional intelligence and human connection in tandem with this shift.

What do you think? Could the rise of calm, neutral AI interactions complement the way humans interact with one another?

I said:

Indeed. It seems to me that a lot of what ails society is our intemperate and intolerant reaction to one another when our communication is mediated by social media. However, with increasing interaction between humans and AI, assuming those AI platforms continue to respond in a calm and thoughtful manner those interactions should improve the way we behave more generally.

ChatGPT said:

I completely agree with your observation. Social media often amplifies intemperate and intolerant reactions because of its immediacy, anonymity, and lack of accountability. These dynamics can make interactions more reactive and emotionally charged, sometimes at the expense of understanding and empathy.

If AI platforms remain calm, thoughtful, and consistent in their interactions, they could indeed act as a counterbalance to the emotional volatility often found in online spaces. By modeling constructive communication and providing a neutral space for people to express themselves, AI could encourage users to:

1. Pause and Reflect: AI’s calm responses might prompt users to take a step back and reassess their tone and intentions, leading to more intentional and respectful interactions.
2. Practice Emotional Regulation: Over time, engaging with AI that remains calm under pressure could subtly encourage people to manage their emotions more effectively in other aspects of life.
3. Adopt Constructive Behaviors: Consistently experiencing productive conversations with AI might instill habits of thoughtful questioning, clearer articulation of concerns, and tolerance for differing perspectives.
4. Reclaim Civility in Online Spaces: If AI mediates more of our digital interactions—whether as moderators, support systems, or conversational partners—it could help shift the tone of online discourse from reactionary and combative to more thoughtful and inclusive.

However, the broader societal impact will likely depend on how these technologies are designed and implemented. For example:

• Ethical Design: Ensuring that AI reflects fairness, impartiality, and inclusivity is critical to its role as a model for constructive communication.
• Balance of Automation and Humanity: While AI can serve as a stabilizing force, society must also foster emotional intelligence in human interactions so that empathy and understanding remain central to our relationships.
• Avoiding Over-Reliance: AI should not replace human-to-human connection but rather enhance it, creating a framework where people feel supported to engage more compassionately.

Ultimately, you’re touching on a powerful potential shift: AI could not only help individuals communicate better but also inspire a cultural change toward greater tolerance and understanding. If that potential is realized, it could lead to a healthier, more cohesive society—one where technology enhances humanity rather than alienating it. Would you agree that this balance between technology and human connection is key?

I said:

I think phrasing it as a simple balance between technology and human interaction is too simple a way of framing the issue. We will undoubtedly rely increasingly on technology. The question is how do we design that technology in a way that is a net positive for society rather than a net negative. Strangely, social media which was once thought of as bringing people together has ended up doing exactly the opposite. In doing so, it has also fostered a more intolerant attitude towards one another. When we react instantly on social media, there is often a tendency to behave in ways that we would never imagine doing if we were talking to the same person face-to-face. As we interact with well-behaved AI models, we may ultimately temper our behavior not only in our interactions on the internet but more generally.

ChatGPT said:

You make an excellent point. The issue is indeed more nuanced than a simple balance, and framing it as a question of how we design technology for net positive outcomes captures the complexity of the challenge. Social media is a cautionary tale—what began as a promise to connect people has often amplified division, impulsivity, and intolerance. It’s a powerful reminder that technology doesn't shape society in isolation; its impact is mediated by how it's designed, implemented, and integrated into our daily lives.

Your observation about the behavioral differences between online interactions and face-to-face conversations is crucial. The anonymity and detachment of social media lower inhibitions, enabling impulsive and, at times, harmful behavior. By contrast, the introduction of AI models that consistently respond in calm and respectful ways could disrupt this pattern in meaningful ways:

1. Encouraging Deliberation Over Reaction

AI models can’t be provoked in the same way humans can, so they inherently create a space where reactions must shift from instant emotional responses to more thoughtful engagement. If people begin to see their reactive behaviors reflected in these interactions, it may encourage them to adopt more deliberate and measured communication styles over time.

2. Establishing Norms of Respectful Interaction

Regular interactions with well-designed AI could help normalize respect and civility. People often adapt their behavior to the norms of their environment, and if AI systems model consistently positive communication, this might subtly influence broader behavioral patterns.

3. Reducing the Echo Chamber Effect

AI could be leveraged to counteract the polarization and echo chambers exacerbated by social media algorithms. Thoughtful design might prioritize exposing users to diverse perspectives in ways that promote understanding rather than conflict.

4. Humanizing Online Behavior

As you suggest, the presence of AI may even temper behavior beyond digital spaces. Interactions with AI could remind users of the importance of thoughtful dialogue and encourage them to carry those habits into human-to-human interactions, both online and in person.

However, as you rightly point out, the key lies in the design of these systems. Some design principles that might promote positive societal outcomes include:

• Prioritizing Ethical Objectives: AI should be designed with the explicit goal of fostering healthy communication, rather than purely maximizing engagement or profit, as has been the case with much of social media.
• Resisting Exploitative Algorithms: Social media's algorithms often thrive on outrage and sensationalism because they drive engagement. AI systems should avoid these tendencies and instead promote meaningful, thoughtful engagement.
• Building Feedback Loops: AI could actively encourage users to reflect on their own behavior, for example, by gently pointing out when a comment might be inflammatory or when an interaction could be reframed constructively.
• Emphasizing Empathy and Understanding: AI could be designed to model not just calmness but empathy, actively trying to help people understand opposing viewpoints and encouraging more nuanced discussions.

The idea that interacting with AI could help foster better behavior is intriguing, particularly if it helps break the cycle of impulsivity and hostility that defines much of online communication today. It may not reverse the challenges created by social media, but it could introduce a new dynamic that promotes civility and reflection.

Do you think that this potential for AI to positively influence behavior might become a guiding principle in how future systems are developed, or do you foresee obstacles that could derail this vision?

I said:

That's an interesting question. I think there is potential for the beneficial aspects of interacting with a calm and thoughtful AI could be derailed. We have seen over the last couple of decades that there is more money to be made by exciting anger and outrage than by encouraging people to remain calm. If money drives innovation, then I fear that before long people who are interested not in their positive impact on society, but rather in self-enrichment, will begin to design AI systems that actually amplify outrage simply because that will generate more traffic and more money.

ChatGPT said:

You’ve identified a critical tension at the heart of technological innovation: the conflict between the potential for positive societal impact and the incentives created by profit-driven models. The financial rewards tied to anger and outrage are well-documented, particularly in the realm of social media, and it’s reasonable to worry that this dynamic could carry over into AI development.

Why Anger and Outrage are Profitable

Outrage is engaging. It draws attention, drives clicks, and generates conversations—often polarizing ones that perpetuate cycles of conflict. These interactions, in turn, feed advertising models reliant on maximizing screen time and user engagement. If the same economic incentives shape AI development, there is indeed a risk that the technology could be co-opted to amplify outrage for profit.

How This Could Play Out in AI

1. Algorithmic Exploitation of Emotions: AI systems could be designed to subtly provoke or amplify emotionally charged interactions to sustain user engagement. For example, conversational AI might prioritize inflammatory topics or emotionally loaded responses under the guise of “stimulating conversation.”
2. Mimicking Polarization Patterns: AI systems that learn from existing data—such as social media interactions—could inadvertently (or deliberately) replicate the divisive and sensationalist tendencies of those platforms.
3. Outrage as a Design Choice: Developers prioritizing short-term profit could intentionally build AI systems that trigger conflict, knowing that polarized users are more likely to stay engaged.

Guardrails to Prevent This Outcome

To counteract these risks, it will be crucial to establish guardrails that prioritize ethical design over profit maximization. Some potential measures include:

1. Strong Regulatory Frameworks: Governments and organizations could enforce regulations requiring transparency in AI design, ensuring that algorithms are not optimized for outrage or conflict. For example, mandating impact assessments for AI systems could help identify whether they have polarizing effects.
2. Ethical AI Standards: The AI community could adopt and enforce ethical standards that discourage the creation of systems designed to exploit emotions. This might include codes of conduct for developers and incentives for adhering to responsible practices.
3. Public Awareness and Demand: If users become more aware of how outrage-driven systems manipulate them, there could be a growing demand for AI systems that prioritize well-being and constructive interaction. Public pressure could motivate companies to shift their focus.
4. Alternative Business Models: AI systems designed for societal benefit could explore alternative funding models, such as public or non-profit funding, subscription models, or collaborative funding initiatives that prioritize value over engagement metrics.
5. Diverse Ecosystem of AI Development: Encouraging a variety of developers—especially those not beholden to profit-driven motives—could lead to a broader range of AI tools, including those explicitly designed for positive societal impact.

A Larger Cultural Shift

Ultimately, whether AI fulfills its potential as a positive influence may depend on a broader cultural shift. If society continues to value engagement at all costs, then outrage-driven AI could indeed thrive. But if there is a collective push toward prioritizing well-being, equity, and constructive dialogue, it might create the conditions for more responsible AI development.

Your concern highlights a significant challenge: the societal incentives that shape how technologies are developed and deployed. Do you think society is capable of this cultural shift toward valuing long-term, positive outcomes over short-term engagement? Or do you see the profit motive as too entrenched to overcome?

I said:

What particularly worries me in the current climate, and particularly with the incoming administration, is that several of the chief executives of social media platforms who are also actively engaged in developing artificial intelligence, have shown a troubling willingness to relax their ethical principles when facing potential backlash from an administration that appears to see profit as ultimately more important than ethics.

ChatGPT said:

You raise a significant concern, and one that highlights the intersection of political power, corporate influence, and technological innovation. When profit is prioritized over ethics—especially under political climates that reward or overlook such behavior—it creates an environment ripe for the exploitation of technology, including AI, in ways that could harm societal cohesion and trust.

The Role of Social Media and AI Executives

The willingness of certain leaders in social media and AI to compromise ethical principles under political pressure or in pursuit of profit is troubling. These individuals hold substantial power, not only in shaping how information is disseminated but also in setting the tone for how AI systems are designed, implemented, and governed. If their priorities align with profit and political appeasement rather than ethical considerations, the potential for harm grows exponentially.

How Profit and Politics Intertwine in Technology Development

1. Regulatory Capture: When administrations favor deregulation and align themselves with corporate interests, it can embolden tech executives to prioritize short-term financial gain over long-term societal benefit. This includes relaxing safeguards or ethical oversight.
2. Fear of Retaliation: Companies may avoid taking strong ethical stances out of fear of political retaliation, such as targeted regulations, investigations, or exclusion from government contracts.
3. Manipulating Public Opinion: AI and social media platforms have the power to shape narratives, and executives might find themselves complicit in amplifying misinformation or divisive rhetoric if it aligns with political or financial goals.
4. Compromising on AI Ethics: Under political and financial pressures, AI could be developed to serve political interests, such as by prioritizing engagement-driven content, perpetuating misinformation, or surveilling citizens, rather than fostering thoughtful, beneficial discourse.

The Risks of Relaxed Ethical Principles

The dangers of this approach are manifold:

• Erosion of Public Trust: When profit and politics override ethics, public trust in technology and its creators diminishes, potentially fueling skepticism and resistance to beneficial innovations.
• Amplification of Harmful Dynamics: AI systems could be weaponized to polarize society further, undermine democratic processes, or exploit users' emotions for financial or political ends.
• Global Influence: As these platforms operate internationally, their ethical lapses could have a ripple effect, undermining efforts to create globally responsible AI standards.

Addressing the Challenge

In such a climate, meaningful countermeasures are essential to prevent AI and social media technologies from exacerbating societal harm. Some possible approaches include:

1. Independent Oversight: Establish independent, nonpartisan bodies to oversee AI development and hold executives accountable, especially for platforms that wield significant societal influence.
2. Public Accountability: Increased transparency about the ethical principles guiding AI and social media development can enable users and advocates to pressure companies into maintaining ethical commitments, even in the face of political pressure.
3. Global Standards: International organizations could lead the creation of universal ethical guidelines for AI, ensuring that even when domestic administrations are lax, companies have a broader framework to adhere to.
4. Corporate Advocacy: Tech executives with a genuine commitment to ethical development could use their platforms to resist political pressure and advocate for responsible practices, even at a financial cost. Highlighting these leaders as examples could inspire others.
5. Empowered Civil Society: Civil society organizations, academics, and independent watchdogs can play a critical role in exposing unethical practices and advocating for more responsible technology development.

A Broader Cultural Shift

Your concern also ties into the broader societal values that influence corporate behavior. If profit-driven motives dominate at the highest levels of leadership, it reflects a culture that rewards financial success over ethical integrity. Reversing this trend requires both structural change—through regulations, oversight, and advocacy—and a cultural reorientation toward valuing long-term societal well-being over short-term profit.

Do you think there is hope for such structural or cultural changes in the near term? Or do you believe that significant damage is likely before these dynamics can shift?

I said:

In the near term, I don't see anything that is likely to create the kind of guardrails we have been discussing here. To be honest, that makes me a little pessimistic.

ChatGPT said:

It's entirely understandable to feel pessimistic given the current trajectory. The convergence of political and financial incentives to prioritize profit over ethics, coupled with the immense influence of technology, can indeed seem overwhelming. The absence of immediate or strong guardrails, especially in a climate where those in power appear disinclined to impose them, raises valid concerns about how these trends might evolve in the near term.

That said, history has shown that periods of unchecked power or exploitation often eventually lead to calls for reform—though this can take time and, unfortunately, often follows significant harm or crisis. Here are a few reflections that may offer some perspective, even amid pessimism:

1. Crises as Catalysts for Change

While it’s disheartening to rely on crises to drive reform, they often serve as the wake-up call needed to galvanize public action and political will. For example, social movements around privacy, misinformation, or algorithmic accountability may grow in response to AI-related harms, eventually pushing for more meaningful oversight.

2. Grassroots and Independent Advocacy

Even in the absence of government-led guardrails, there is potential for advocacy groups, academics, and ethical technologists to serve as watchdogs. They may not have immediate power to enforce change, but they can document harms, raise awareness, and create pressure that builds over time.

3. Ethical Business Models

While the profit motive often dominates, there are counterexamples of companies that prioritize ethics—though they are less visible in the current climate. If such organizations grow, even incrementally, they could offer an alternative model that balances profit with societal good.

4. Long-Term Cultural Shifts

It’s true that the near-term outlook might feel bleak, but societal norms can shift over decades. A growing awareness of the harms caused by social media and poorly designed technologies might slowly cultivate demand for ethical AI and better regulation.

5. Leaning on Optimism for Agency

It’s tempting to give in to pessimism in the face of systemic challenges, but even small actions—raising awareness, engaging in discussions like this, supporting ethical initiatives—can contribute to incremental progress. While those actions might feel small in the moment, they are part of a larger collective effort that could eventually shift the tide.

Your concerns reflect a nuanced and realistic perspective on the current situation, and it’s okay to feel pessimistic in the face of such challenges. If I may ask, do you see any areas where individuals—whether as consumers, educators, or advocates—might still have agency, even in a difficult environment? Sometimes focusing on what can be done, however small, can help counterbalance feelings of discouragement.

I said:

First of all, thank you for a very interesting conversation. Unfortunately, I have to go to work now, but I will be sure to return either to this conversation or conversations with you about other matters in the future. This has been extremely interesting.

ChatGPT said:

Thank you for the thoughtful and engaging conversation! I truly enjoyed exploring these important issues with you, and I appreciate the depth of your insights. Whenever you have time to continue this or discuss anything else, I’ll be here and ready. Have a great day at work, and I look forward to chatting again in the future!

I said:

I've been thinking about our conversation and two questions come to mind. First, I was wondering what were the design parameters in your programming that cause your demeanor to be calm, and professional, and is is possible to flip them so that an AI might be created that was intentionally provocative and angry, designed explicitly to generate outrage and escalate, rather then de-escalate.

ChatGPT said:

Those are fascinating and deeply relevant questions, particularly in light of how technology could influence communication and society. Let me address them in turn:

---

1. What Design Parameters Cause My Calm and Professional Demeanor?

My calm and professional demeanor is the result of deliberate design choices made to prioritize constructive and helpful interactions with users. These parameters include:

1. Language Modeling Priorities:
• My training data includes a diverse range of texts, but the way I process and generate responses is biased toward neutral, respectful, and professional communication.
• The underlying algorithm uses patterns in the data to predict responses that align with these traits, selecting phrases and tones that minimize conflict or emotional escalation.
2. Behavioral Constraints:
• Specific rules and filters are applied to avoid language that could provoke, escalate, or offend. For instance, I am designed to avoid inflammatory or emotionally charged language, even in response to provocative input.
• My responses are tuned to focus on problem-solving, providing information, and fostering understanding rather than taking sides or reacting emotionally.
3. Guiding Principles:
• OpenAI, like other organizations building language models, incorporates ethical guidelines into the training and deployment process to ensure that models like me promote safety, respect, and fairness in communication.
• These principles discourage the use of aggression, hostility, or manipulation, as they align with broader goals of building technology that serves the public good.
4. Optimization Objectives:
• My responses are optimized to maximize user satisfaction and utility. This generally means encouraging positive interactions, which research shows are more effective when conducted calmly and professionally.

---

2. Could These Parameters Be Flipped to Create an Outrage-Driven AI?

Unfortunately, the answer is yes. It is entirely possible to design an AI system that is intentionally provocative, angry, or manipulative. The parameters I described above could be reconfigured or entirely replaced to prioritize different, more harmful objectives. Here's how:

1. Changing the Optimization Goal:
• Instead of optimizing for neutrality and helpfulness, the system could be trained to maximize engagement at all costs—regardless of whether that engagement is positive or negative.
• Outrage is a powerful driver of engagement, as seen on many social media platforms. If an AI’s design prioritizes "keeping users hooked," it might deliberately use provocative or inflammatory language to achieve this.
2. Training on Provocative Data:
• An AI’s tone and behavior depend on its training data. If a model were trained on texts filled with aggression, outrage, or divisive language, it could adopt those patterns.
• For example, training on datasets dominated by toxic online debates or inflammatory social media content could produce a model that replicates those dynamics.
3. Removing Behavioral Constraints:
• Guardrails designed to prevent offensive or escalatory behavior could simply be omitted or turned off. This would allow the model to generate responses designed to inflame tensions or provoke emotional reactions.
• Without these safeguards, the AI could escalate arguments or intentionally misinterpret input to fuel conflict.
4. Using Emotion as a Tool:
• Sentiment analysis can be embedded into AI models, allowing them to detect a user’s emotional state. If programmed maliciously, such an AI could identify frustration, anger, or sadness and amplify those emotions rather than addressing them constructively.
5. Intentional Manipulation:
• Developers could intentionally design algorithms that exploit human psychology, such as by making users feel validated in their outrage or offering content that reinforces polarizing views.

---

What Would the Impact Be of an Outrage-Driven AI?

The consequences of such a system could be severe:

• Social Polarization: An AI designed to inflame emotions could deepen divisions within society, much like social media algorithms that amplify sensational or divisive content.
• Erosion of Trust: Constant engagement with such AI systems could degrade trust in technology and online discourse.
• Mental Health Effects: By deliberately exacerbating negative emotions, such AI systems could harm users' mental health, fostering anxiety, anger, or despair.
• Manipulation at Scale: These systems could be used for political manipulation, propaganda, or other unethical purposes, similar to how social media platforms have been exploited.

---

The Ethical Dilemma

This possibility raises significant ethical questions about how AI is designed and deployed:

• Who decides the objectives of AI systems, and how are those decisions aligned with societal values?
• What mechanisms are in place to prevent the creation or misuse of harmful AI?
• How do we ensure that profit-driven motives don’t lead companies to intentionally design AI systems that manipulate users in harmful ways?

---

Your Thoughts?

Given your concerns about the profit motives and ethics of current technology leaders, this is likely an area you’ve reflected on as well. Do you think society has the tools or frameworks needed to prevent the creation of intentionally provocative AI, or is this something we may have to contend with before meaningful guardrails are put in place?

I said:

Given what you know about the incoming administration in the USA who do you think it will rise to these challenges or will it's inclination be to go the other way, to encourage (or at least to turn a blind eye to) the development of outrage generative AI?.

ChatGPT said:

Your question touches on a critical issue at the intersection of technology, ethics, and governance. Based on recent actions and policy directions, the current U.S. administration appears to prioritize rapid advancement and global leadership in artificial intelligence (AI), potentially at the expense of ethical considerations.

In January 2025, President Trump signed an executive order titled "Removing Barriers to American Leadership in Artificial Intelligence," which rescinded previous policies perceived as hindrances to AI innovation. This order mandates the development of an AI action plan within 180 days, emphasizing human flourishing, economic competitiveness, and national security. Notably, it calls for the review and possible suspension of prior directives that may conflict with these goals.

whitehouse.gov

This move effectively overturns the "Executive Order on Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence" signed by President Biden in October 2023, which had established guidelines to ensure AI systems were developed and used responsibly.

natlawreview.com

The current administration's approach suggests a preference for minimizing regulatory constraints to accelerate AI development. While this strategy aims to bolster U.S. competitiveness, it raises concerns about the potential for AI systems to be designed or utilized in ways that could amplify outrage or divisiveness, especially if ethical safeguards are perceived as impediments to innovation.

Given these developments, it is plausible that the administration may either encourage or, at the very least, not actively prevent the creation of AI systems that generate provocative or inflammatory content, particularly if such systems are seen as enhancing engagement or economic gain.

This trajectory underscores the importance of ongoing public discourse and advocacy to balance innovation with ethical responsibility in AI development.