Philosopher & AI Ethicist @GoogleDeepMind · @LeverhulmeCFI @Cambridge_Uni | Consciousness, Machine Minds, AGI, Human-AI Relationships | All views my own
Cambridge, England
- Tweets 19,110
- Following 9,658
- Followers 38,586
- Likes 134,068
Pinned Tweet
After years of bumping against Twitter's character limit, I've finally started a blog, called Polytropolis (a Homeric pun). As my first post, I'm sharing my Berggruen-shortlisted essay on why the public will decide AI is conscious before scientists do.
polytropolis.com/p/behaviour…
62
51
398
140,071
My 9 albums. Mostly things that hit me at formative moments; the album format also favours some genres over others. My taste today is more eclectic and frankly bizarre (a representative selection would need way more K-pop, sea shanties, chiptune and show tunes).
6
32
1,225
I disagree with Matt Walsh and Megan here.
(1) We don’t know how to build general-purpose AI without any anthropomimetic traits
(2) Anthropomimetic AI has lots of valuable uses
(3) Some problems (eg sycophancy) reflect AI being trained to be *less* humanlike rather than more so
I never take any pleasure in boosting Matt Walsh but he is 100% correct here. The anthropomorphic form is a result of intentional decisions by the human designers, there is no good reason for it, and it is nefarious. People should be talking about that more.
42
17
232
7,558
Even if AIs were conscious, it wouldn’t follow that our current treatment harms them. Meanwhile, we inflict industrial-scale horrors on animals whose capacity to suffer we scarcely dispute. Recognising consciousness clearly doesn’t ensure taking welfare seriously
The discussion about AI consciousness is one of profound unseriousness. If we truly took it to be conscious, even as conscious as a frog, the way we should treat each instance would have to change so dramatically that the labs would have to shut down. Everything else is verbal gymnastics.
46
62
545
16,245
(As a sidenote I have a piece on our conflicting moral attitudes towards consciousness, animals, and AI coming out in @bigthink some time soon)
3
35
1,068
After a stressful week, I like to centre myself through the Finnish practice of kalsarikännit, which roughly translates as “taking time to reconnect with oneself and honour one’s needs”.
22
4
137
5,069
Me drawing maps at school in the back of my workbook instead of paying attention, aged 5–18
5
1
37
2,442
Protip: if you’re ever having a sauna and it’s getting a bit crowded, try explaining enthusiastically to your fellow saunees how perception, intelligence and even life itself can all be modelled as homeostatic control systems, you’ll be amazed how quickly it thins out
38
14
311
7,822
On a serious note, my top advice for anyone devising a theory of consciousness (besides “don’t”): learn the existing landscape and where your view fits. Which theories are closest? What does yours improve on? How does it handle the standard objections to those theories?
48
14
230
21,885
these really are the geocities days of AI, we should remember to enjoy them before everything gets too slick
40
35
519
11,455
One crazy thing about loving someone profoundly is that you overfit your affection to every part of them. Even their worst traits become hard to criticise, because without them they’d be less than 100% themselves.
23
12
273
7,263
Henry Shevlin retweeted
This piece argues that EA is bad as a totalizing worldview. I agree, and I wish EAs had more uncertainty and pluralism. But critics should see that EA's method is about problems *at current margins.* i.e. we torture too many chickens now, so we should use *some* resources to stop
Effective altruism is admirable, but dangerous. As the movement has grown stronger, it has become stranger bit.ly/47th7SZ
1
9
103
5,397
Very grateful to my son for introducing me to Undertale & Deltarune. Just finishing Chapter 5 and increasingly convinced that Toby Fox is one of the great composers of our time who happens to distribute his music in brilliant games about annoying dogs.
piped.video/1kGmc1BOyPY?is=eQT0…
11
3
152
4,939