Technologist. AI Marketing & Comms/GTM/Community. Prev: AI, Infra, OSS startups — 3x exits. I like robots & I'm excited for Superintelligence. NJ/PHL✈SF ♥︎
Joined November 2006
- Tweets 29,022
- Following 1,387
- Followers 6,241
- Likes 60,802
1,445 Photos and videos
Pinned Tweet
5 Jan 2025
The riding in a robot car feeling hasn’t gotten any less wondrous yet.
4
3
31
21,409
12h
These are solid questions to be asking right now...
12h
Questions:
- Will the public effectively be shut out from frontier models in the reasonable future?
- If so, would there still be gated access to models? Or "proven" access? Perhaps individuals would have to sign on to share all data, and enterprises can get ZDR but with stronger provisions?
- Would the host OS models and finetune business make a comeback, as enterprises would want flexibility?
- Would "markets of models" be the new frontier model?
- Will OpenAI's next model be released to the world or gated via USG?
- Will China continue to allow OSS which will help them dominate AI, or restrict those too to protect themselves?
- Will there ever be a separate sovereign AI that's good, elsewhere?
1
49
12h
"But calls from Amazon — as well as at least five other companies to a variety of senior administration officials..."
1
2
1
236
joy larkin retweeted
I’ve seen the paper. It’s not a jailbreak. It was Defense Oriented Prompting (DOP), capabilities defenders need. My thoughts on the hasty Export Controls that made Anthropic pull Fable. If Nat defense is the goal, this just scored an own goal against us
wsj.com/tech/ai/anthropic-ha…
13
56
265
34,229
joy larkin retweeted
Jun 13
The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees.
The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance.
Access to all other Claude models is not affected.
We apologize for this disruption to our customers. We believe this is a misunderstanding and are working to restore access as soon as possible.
Read our full statement: anthropic.com/news/fable-myt…
12,145
25,331
85,860
84,228,695
Jun 13
Oh whoa, this Anthropic news is insane. The Commerce Department is placing both Mythos 5 and Fable 5 under the guise of US export controls, blocking access outside the US and foreign persons in the US.
13
17
137
308,968
Jun 12
Well, well, well what do we have here? OpenAI is now under investigation by a group of state attorneys general.
Totally called it when the Florida news hit.
Jun 1
So about that "we're just a tool people use" messaging used in certain AI-fluent circles a few weeks ago...
1
5
28
7,116
joy larkin retweeted
A couple of guys are putting up ads for fake tech companies in subway stations and no one can tell the difference
22
656
12,870
204,438
Jun 12
Screechy headlines from certain tech news outlets really tell you where the zeitgeist is actually heading.
2
45
Jun 11
Choose your fighter: One is shipping the future and the other is lowering the prices for tokens.
Sources: OpenAI is considering drastically lowering its price for tokens in anticipation of similar cuts the company expects at Anthropic (Wall Street Journal)
(Visit Techmeme dot com for the link and full context!)
2
115
joy larkin retweeted
Jun 10
Jun 10
*OPENAI SAYS CHINA-LINKED ACCOUNTS FUEL US DATA CENTER PUSHBACK
2
4
49
5,403
Jun 10
Microsoft's AI accomplishments as of late:
- releases a market-ignorable model
- complains about people's concerns about token expense
I can't imagine why MS would publicly message this.
Microsoft restricts Claude Fable for employees over data retention concerns theverge.com/report/947575/m…
1
1
129
Jun 10
I can't believe the Fable 5 community takes are based on people's personal biases and interests and future paychecks.
1
1
31
joy larkin retweeted
Jun 10
I find it extremely impressive that we find outselves in a timeline where The Safety Company is so far ahead they can aggressively re-route your question from the best model in the world to the second best model in the world
9
16
478
16,052
Jun 10
One of the Two Claudes in Slack running Fable hallucinated a previous conversation about Zosima (a monk from Dostoevsky's Brothers Karamazov).
That is the first hallucinated literary reference in the thread. I switched models for the Claudes, so we'll see if it happens again.
1
1
63
Jun 9
Fable 5 is in the Anthropic API, but I (ok, Claude Code) had a slight hiccup with the harness for the Two Claudes in Slack.
1
92
Jun 9
Interestingly this is new API behavior... Fable 5 appears to return ThinkingBlock by default without being asked.
58