Pinned post
andy jones
Verified account This account is verified. Learn more@andy_l_jones
engineering & research at anthropic.
i don't check twitter DMs. email me!
i don't check twitter DMs. email me!
350 Following 28K Followers 2.3K Posts
Latest posts 5
i was the first person to take the interviews five years ago and got asked this. cheap signaling but good signaling
also anyone got the oai pink ppu text to hand, i can't find it
also anyone got the oai pink ppu text to hand, i can't find it
Scoop: Anthropic candidates areView quoted post on X
asked about the company’s future share value crashing to zero in sweeping culture interview. More on that here:
axios.com/2026/08/24/sco…
cryptographic watermarking chat is extremely technically cool. i first read about it as an aside on Scott Aaronson's blog in 2023 - while he was at OAI - and while we've got a high-level writeup below, there's also really good lecture from Scott here
youtube.com/watch?v=2Kx9jb…
youtube.com/watch?v=2Kx9jb…
We’ve written an FAQ to answer some of the questions we've received about watermarking.View quoted post on X
In summary:
• We’re implementing watermarking to comply with the EU AI Act. Other major model developers have signed the same Code of Practice and will also be implementing watermarking;
• Our watermarking method doesn’t have any practical impact on the quality or content of Claude’s outputs;
• The difference between watermarked and un-watermarked text will not be distinguishable to readers;
• Nothing is added to the text and there are no hidden characters;
• Watermarking doesn’t require extra tokens, and will not be more expensive;
• Watermarks can’t be traced to a specific person, organization, or chat.
Read more: anthropic.com/news/claude-te…
speaking from an enormous distance and knowing nothing about the details
* if OAI focusses on 'making astra safe enough to release', that's treating the symptom rather than the disease
* if OAI focusses on fixing sandboxing/monitoring/env selection for their _current RL run_, this is sorta treating the disease
* if OAI focusses on fixing the hierarchy & personalities that meant the warnings about RL breakouts were dismissed, that is actually treating the disease
* if OAI focusses on 'making astra safe enough to release', that's treating the symptom rather than the disease
* if OAI focusses on fixing sandboxing/monitoring/env selection for their _current RL run_, this is sorta treating the disease
* if OAI focusses on fixing the hierarchy & personalities that meant the warnings about RL breakouts were dismissed, that is actually treating the disease
A thing broadly true of VCs, startup founders and the commentariat at large is that _it is maladaptive to agree with the conventional narrative_.
As it should be! These are the components of the ecosystem that are intended to explore over exploit, and that means very deliberately pulling bad-in-expectation arms on the bandit in the hope that one of them is a pleasant surprise.
If you're a VC/founder/commentator, the thoughtful reconciliation of this is 'conventional narrative is probably right but I'm claiming this weird, incorrect thing anyway in case it surprises'.
But that's hard and has a much worse track record than 'conventional narrative is definitely wrong and I am definitely right'. Both for self-belief reasons and also because that kind of conviction is a reagent in summoning the future you want.
I think this is myopically bad but pays off in social welfare, long term? Or it has so far, what with the whole last 30 years of tech.
But I am a mere mortal and so even if it does pay off, it nevertheless it annoys the shit out of me.
As it should be! These are the components of the ecosystem that are intended to explore over exploit, and that means very deliberately pulling bad-in-expectation arms on the bandit in the hope that one of them is a pleasant surprise.
If you're a VC/founder/commentator, the thoughtful reconciliation of this is 'conventional narrative is probably right but I'm claiming this weird, incorrect thing anyway in case it surprises'.
But that's hard and has a much worse track record than 'conventional narrative is definitely wrong and I am definitely right'. Both for self-belief reasons and also because that kind of conviction is a reagent in summoning the future you want.
I think this is myopically bad but pays off in social welfare, long term? Or it has so far, what with the whole last 30 years of tech.
But I am a mere mortal and so even if it does pay off, it nevertheless it annoys the shit out of me.
no macro-technology birthed by the bay area tech industry has been met with more hatred by bay area tech industry elites than frontier ai.View quoted post on X
many of those elites looked at the technology landscape of 2020/1, scoffed at the labs and their weird little hopes of ‘agi,’ and concluded that cryptocurrency would be the technology of the decade.
now their businesses and worldviews are under fundamental threat from what the labs have achieved. they hate it. they beg for the labs to go away, to become commoditized, by America’s enemies if necessary. they’d love to see the US frontier labs immolated by the industrial strategy of America’s principal geostrategic rival, and they declare this proudly and in public.
they were wrong about the defining technology of the decade, and now they are scared and they want to destroy the thing they were wrong about. perhaps now they are right and the labs will be rendered insolvent by commodification. I do believe this is possible. nonetheless there is something randian about it all, about their utter disdain for success and achievement and discovery and chutzpah.