Former antitrust litigator; now writing for Attention Is and senior fellow at The Midas Project. My views are my own. ~ 🪶 be nice to your AIs ~
Joined June 2026
- Tweets1.7K
- Following1K
- Followers210
- Likes6.5K
Pinned Tweet
I like reading (and listening!) to books. You should too. There’s a lot to be learned and a lot of new worlds to discover. Currently reading:
zzzSam retweeted
Replicated this a few minutes ago.
Google AI gets cagey when asked who won the 2020 elections arstechnica.com/google/2026/…
And see, when I read Sundar’s post, all I can think of is how the companies are now actively trying to develop the record of how they are “not” pacing: teapawt.net/sundarpichai/status/21…
zzzSam retweeted
"Carbon shows the pace at which we’re pushing the frontier"
Lol @ the labs already appropriating 'Pacing The Frontier' to mean capabilities acceleration.
I did say we'd be releasing our models periodically...introducing Gemini 4 Carbon, our next frontier model.
Following Argon's release just two weeks ago, Carbon shows the pace at which we’re pushing the frontier, it ranks #1 on the recently updated v5.0 of @ArtificialAnlys Intelligence Index at 59. It also leads across key cybersecurity evaluations scoring 88.2% on SEC-bench Pro for finding vulnerabilities, 54.8% on ExploitGym for exploitation, and 71.0% on CWE-bench for patching.
We’re seeing significant improvements in long-horizon tasks even compared to Argon, and we’ve strengthened frontier safeguards.
As we did with Argon, we’re going to roll it out gradually + responsibly. Some trusted cyber defenders will get access starting this week. We'll be expanding it to more partners through our Fairwind Program, ahead of wider availability soon, so hang tight!
Governments should use their contracts with the big AI developers to extract safety commitments and standards. I'm just sayin...
We're expanding our partnership with @USGSA. Under a new OneGov agreement, we are bringing Claude to federal, state, local, tribal, and territorial governments at 40% off.
anthropic.com/news/expanding…
We'll see what the companies say in their Motion(s) to Dismiss... FYI, Motion(s) are due December 11. Mark your calendars. It's gonna be super duper fun to read about how they are all still racing... 🫠
zzzSam retweeted
Anthropic is saying that AI models could pose catastrophic or existential risks to humanity.
At the same time, Anthropic is planning the largest IPO in history, which would accelerate the development of this technology while leaving Americans holding the bag if the AI bubble bursts.
I wrote to CEO Dario Amodei asking for real answers about the risks before they try and go public.
zzzSam retweeted
@arb8020 notices that the GPT-5.5 Codex system prompt says to stop talking about goblins
teapawt.net/arb8020/status/2048958…
gpt-5.5 prompt for codex seems to have a duplicated line trying to get it to not talk about creatures?
Never talk about goblins, gremlins, raccoons, trolls, ogres, pigeons, or other animals or creatures unless it is absolutely and unambiguously relevant to the user's query.
[...]
Never talk about goblins, gremlins, raccoons, trolls, ogres, pigeons, or other animals or creatures unless it is absolutely and unambiguously relevant to the user's query
gh link:
github.com/openai/codex/blob…
zzzSam retweeted
New Anthropic research: How AI agents shape experts' judgment calls
When experts delegate work to an agent, do they still make the key decisions? We tested this in a randomized trial with 50 professional data analysts.
zzzSam retweeted
Do you like Twitter but also hate Twitter? I've now developed an entirely autonomous Substack of the best tweets every day, with tweets selected and assembled entirely by LLMs. thequotetweet.substack.com/
zzzSam retweeted
When Meta says they’re going to release a safety report “soon”, how soon do they mean?
It’s been over two weeks since they released Muse Wonder, and they still haven’t released it.
On Oct. 12, Meta released Muse Wonder with only a 13-page model card and a promise to release a safety report “soon.” 11 days later, Meta updated the model card. There’s still no safety report. 🧵
ALT Before and after of Meta’s Muse Wonder model card: ExploitGym result changed from 92 to 20, and the original scoring note was replaced with an Oct. 23 update note.