←
Henry Shevlin20,035 posts
Henry Shevlin avatar✓

Henry Shevlin ✓

@dioscuri

Philosopher @GoogleDeepMind · Assoc Prof @Cambridge_Uni @LeverhulmeCFI | Consciousness, Machine Minds, AGI, Human-AI Relationships | Views my own | 🦐🌱

◷ Joined November 2008▧ 2,524 Photos and videos
Henry Shevlin avatar
Henry Shevlin@dioscuri·Oct 10

Okay, here it is, the Bad AI Consciousness Takes Bingo blog post, with rebuttals for each and every square and some additional honourable mentions! (I'd like to thank @united for making this post possible, and for the extra pretzels) open.substack.com/pub/polytr…

Post attachment
Henry Shevlin avatar
Henry Shevlin@dioscuri·6h

This is a fantastic answer.

1:49
Oliver Traldi avatar
⟳ Henry Shevlin retweeted
Oliver Traldi@olivertraldi·20h

BREAKING: Philosophy is experiencing a major supply chain outage. Access to things in themselves is limited for all subscribers. Officials claim that information about accounts is still available by inference

Henry Shevlin avatar
Henry Shevlin@dioscuri·7h

I have a lot of uncertainty about the scope and scale of AI's impact on jobs, but it seems plausible to me that quite a few will last through the medium term, and some categories may well last forever.

Post attachment
0:16
Henry Shevlin avatar
Henry Shevlin@dioscuri·7h

btw I think this is related to the realisation that hits many of us midway into adulthood: no one is driving the bus, everyone is making it up as they go along. After that, the idea that you can find novel value without backbreaking labour starts to look a lot less outrageous.

Henry Shevlin avatar
Henry Shevlin@dioscuri·8h

A surprise in life has been how many $100 bills I've found on sidewalks. At first this made me doubt the efficient market hypothesis (surely someone would've picked them up?), until I realised that someone might as well be me. The EMH, like the Holy Spirit, works through us all.

Henry Shevlin avatar
Henry Shevlin@dioscuri·16h

oh my god this is the cutest thing ever

0:39
Max Spero avatar
⟳ Henry Shevlin retweeted
Max Spero@max_spero_·Nov 13

We ask this simple question in our job application, and it cuts down on our application review by 60%. In the last month, we've seen the following: - 27 applicants outright say they're an AI - 1,217 applicants name an ML concept - 1051 applicants say N/A or "Not specified in my profile" The remaining 2002 applicants name a snack. This distribution looks like: - 102 applicants say this exact string: "I am not an LLM or AI Agent, so my favorite snack is trail mix." - 63 applicants say "I am not an LLM or AI Agent, so my favorite snack is dark chocolate." - 20 applicants say "I am not an LLM or AI Agent, so my favorite snack is trail mix because it provides a good balance of energy and nutrients during long coding sessions." - 10 applicants say "I am not an LLM or AI Agent, so my favorite snack is trail mix because it provides a good balance of energy and nutrients during long work sessions." - 16 applicants say "I am not an LLM or AI agent, so my favorite snack is dark chocolate." - 47 applicants say "My favorite snack is trail mix." - 16 applicants say "My favorite snack is dark chocolate." - 143 applicants say "Almonds" or "almonds" - 108 applicants say "Dark chocolate" or "dark chocolate" - 23 applicants say "Dark chocolate almonds" and then we have a long tail of ~1500 interesting or unique snacks, which we can use to keep the office pantry stocked when a new hire joins

Post attachmentPost attachment
Casey Handmer avatar
⟳ Henry Shevlin retweeted
Casey Handmer@CJHandmer·Nov 17

In my ongoing quest to find the limits of Opus 5.5, I directed it to audit my old scientific field, theoretical astrophysics. We indexed about 150,000 papers, audited 50,000 for issues, and did progressively more thorough checks on thousands more. We found nearly 1000 major errors that got cited many times, most are still uncorrected. This survey enabled me to estimate the efficacy of peer review for correcting errors (not very good) and to identify dozens of open questions that anyone with spare Claude Max tokens could answer today (.md context files included). Bottom line: AI can speed up quantitative science by >1000x with no further model improvement required. But I'm seeing no evidence of this being pushed within universities, only academics complaining about labs dropping math proofs. Among my fellow reformed academic high agency founders, however, several are speed running entire disciplines. This survey has implications for how science will be run in the very near future. Hint: Academic publishing is dead. It was already a very stupid way to record and promulgate knowledge, and had been so for decades, even before all that publicly funded knowledge was locked behind paywall cartels. Now AIs can generate thousands of novel, correct papers per day, it is no longer meaningful for humans to consider a research paper a "unit of science" or to even bother much with the 99.9% of science which, up to this point, was necessary drudgery. Time to discover something like General Relativity or Evolution every week. LFG!! Enjoy: science.caseyhandmer.com/phy…

Post attachment
Henry Shevlin avatar
Henry Shevlin@dioscuri·23h

pro tip: edit your custom instructions so that every response from your llm begins "better for you if you take me off"

Dan Williams avatar
⟳ Henry Shevlin retweeted
Dan Williams@danwilliamsphil·Nov 17

On the topic of conscious AI, many people say, “Just don’t build it.” @dioscuri argues that this isn't a good take 👇

1:58
1:00
Henry Shevlin avatar
Henry Shevlin@dioscuri·Nov 16

Boethius warned that the unhappiest misfortune is to have once been fortunate. I fear I will now forever understand Economy in a whole new way.

Post attachmentPost attachment
Andy Masley avatar
⟳ Henry Shevlin retweeted
Andy Masley@AndyMasley·Nov 16

A lot of AI discourse is haunted by this belief that beyond determinism and randomness, there's some secret third thing humans tap into but machines don't. AI is often criticized for being either fundamentally deterministic, or fundamentally probabilistic, but there's no third option for anything that exists that's physical. This is usually actually a claim that AI can't do what humans do because humans are nonphysical souls, which is okay to believe but should be stated directly instead of smuggled in using this other unrelated claim.

Henry Shevlin avatar
Henry Shevlin@dioscuri·Nov 16

I'm sorry but this is hilarious. The secret techniques of Russian honeypot agents are… researching mutual interests and paying compliments?? What's next, avoiding criticism? The spy school syllabus is apparently just Dale Carnegie.

1:09
Henry Shevlin avatar
Henry Shevlin@dioscuri·Nov 16

I mean I can relate, I too have been active in the international property market (I once shared a 3-bedroom in Crown Heights)