Wednesday 24 May, 2023
Passed this in a corridor yesterday and was struck by it. Extraordinary plants, orchids.
Quote of the Day
“Satire is a sort of glass, wherein beholders do generally discover everybody’s face but their own.”
Jonathan Swift, 1704.
Musical alternative to the morning’s radio news
Green Day | I Fought The Law
One of my favourite songs. Lots of other venerable recordings of it — by The Clash, for example, The Grateful Dead and, first of all (I think), Buddy Holly.
Long Read of the Day
Just calm down about GPT-4 and stop confusing performance with competence.
IEEE Spectrum, an entirely sensible publication, has a great interview with Rodney Brooks on the current feeding frenzy about ‘AI’.
It’s a 13-minute read and worth it.
You wrote a famous article in 2017, “The Seven Deadly Sins of AI Prediction.“ You said then that you wanted an artificial general intelligence to exist—in fact, you said it had always been your personal motivation for working in robotics and AI. But you also said that AGI research wasn’t doing very well at that time at solving the basic problems that had remained intractable for 50 years. My impression now is that you do not think the emergence of GPT-4 and other large language models means that an AGI will be possible within a decade or so.
Rodney Brooks: You’re exactly right. And by the way, GPT-3.5 guessed right—I asked it about me, and it said I was a skeptic about it. But that doesn’t make it an AGI.
The large language models are a little surprising. I’ll give you that. And I think what they say, interestingly, is how much of our language is very much rote, R-O-T-E, rather than generated directly, because it can be collapsed down to this set of parameters. But in that “Seven Deadly Sins” article, I said that one of the deadly sins was how we humans mistake performance for competence.
If I can just expand on that a little. When we see a person with some level performance at some intellectual thing, like describing what’s in a picture, for instance, from that performance, we can generalize about their competence in the area they’re talking about. And we’re really good at that. Evolutionarily, it’s something that we ought to be able to do. We see a person do something, and we know what else they can do, and we can make a judgement quickly. But our models for generalizing from a performance to a competence don’t apply to AI systems.
The example I used at the time was, I think it was a Google program labeling an image of people playing Frisbee in the park. And if a person says, “Oh, that’s a person playing Frisbee in the park,” you would assume you could ask him a question, like, “Can you eat a Frisbee?” And they would know, of course not; it’s made of plastic. You’d just expect they’d have that competence. That they would know the answer to the question, “Can you play Frisbee in a snowstorm? Or, how far can a person throw a Frisbee? Can they throw it 10 miles? Can they only throw it 10 centimeters?” You’d expect all that competence from that one piece of performance: a person saying, “That’s a picture of people playing Frisbee in the park.”
We don’t get that same level of competence from the performance of a large language model. When you poke it, you find that it doesn’t have the logical inference that it may have seemed to have in its first answer…
Do read it. Well-informed common sense from a real expert. And a good antidote to some of the current nonsense about ‘AIs’ that lack intelligence.
What we’ve just lost
Martin Amis has passed away, and the literary world is busy trying to assess his worth. There are already tons of obits and tributes — like a particularly good one by Lisa Allardice in the Guardian.
But I think the best way of understanding what we’ve list is to read him when he’s on song. Like in this 1998 essay on the genius of Jane Austen (and what the movie and TV adaptations get wrong).
Here’s how it begins…
Jane Austen, as they might say in Los Angeles, is suddenly hotter than Quentin Tarantino. But before we try to establish what the Austen phenomenon is, let us first establish what it is not.
About 18 months ago (in the summer of 1996) I went to see Four Weddings and a Funeral at a North London cineplex. Very soon I was filled with a yearning to be doing something else (for example, standing at a bus stop in the rain); and under normal circumstances I would have walked out after ten or fifteen minutes. But these weren’t normal circumstances. Beside me sat Salman Rushdie. For various reasons—various security reasons—we had to stay. Thus Ayatollah Khomeini had condemned me to sit through Four Weddings and a Funeral; and no Iranian torturer could have elicited a greater variety of winces and inches, of pleadings and whimperings. So one was obliged to submit, and to absorb a few social lessons.
It felt like a reversal of the Charles Addams cartoon: I sat there, thoroughly aghast, while everyone around me (save the author of The Satanic Verses) giggled and gurgled, positively hugging themselves with the deliciousness of it all. The only good bit came when you realized that the titular funeral would be dedicated to Simon Callow. I clenched my fist and said yes. No particular disrespect to Simon Callow—but at least one of them was going to die.
“Well,” I said, when it was over, “that was bottomlessly horrible. Why is it so popular?”
“Because,” said Salman, “the world has bad taste. Didn’t you know that?”
Still, “bad taste,” all by itself, won’t quite answer. I can see that the upper classes might enjoy watching the upper classes portrayed with such whimsical fondness. But why should it appeal to 400 plebs from Hendon? In any postwar decade other than the present one, Four Weddings would have provoked nothing but incredulous disgust. A 1960s audience would have wrecked the cinema. Yet now it seems that the old grievances have evaporated, and “the million,” as Hamlet called them, feel free to root for the (congenital) millionaires. They can lapse into a forgetful toadyism, and abase themselves before their historical oppressors…
Blissful. All that talent wrapped up in a diminutive person. May he rest in peace.