Technologist. AI Marketing & Comms/GTM/Community. Prev: AI, Infra, OSS startups — 3x exits. I like robots & I'm excited for safe superintelligence. NJ/PHL✈SF ♥︎
SF Bay Area
- Tweets 29,758
- Following 1,437
- Followers 6,351
- Likes 64,939
👀
Timeline of emergent capabilities
(I had linked to my bad slide version of this 5 times, so I obeyed my inner @gwern and wrote it up)
1
95
If anything, you can call the World Labs to AMD deal the first big neolab acquisition.
Chipmaker AMD agreed to acquire World Labs for $8.2 billion, gaining an artificial intelligence startup founded by industry pioneer and researcher Fei-Fei Li. bloomberg.com/news/articles/…
1
1
22
2,022
All I see are more logos than the last AI security thing they tried promoting...and when that project received little industry traction, it was quietly handed off to another organization.
Today, with over 100 industry partners, we introduced the NVIDIA Open Agent Safety Platform, bringing together OpenShell and Sentry.
Artificial intelligence is extraordinary technology that will advance discovery, productivity, security, health, and prosperity for generations to come.
But its full promise can only be realized when people have confidence that AI is being built to be safe and deployed with wisdom and responsibility.
This is bigger than a single product. It's the beginning of an open ecosystem to build the trust layer for safe agent systems.
Together, we are building the foundation of the AI economy.
Trust and innovation are not in conflict. Safety is how trust is earned. We must build not only the most capable AI, but the most trusted AI, so that this extraordinary technology can realize its enormous promise for the world. nvda.ws/4hcoq7m
1
1
4
325
That OpenAI agent security person's Twitter essay reminded me of SBF's court testimony for some reason.
100
I was working with Claude Opus 5.5 on a personal design project and one of the versions Claude gave me unsolicited was a riff of a Joy Division album cover.
Can't tell me Opus 5.5 isn't advanced intelligence. 😭
1
6
300
Late Friday evening misalignment disclosures announced via lab employee personal accounts should not be the norm. That is no bueno. The pushback becomes personal.
1
3
148
joy larkin retweeted
"why would AI want to kill everyone" *opens this app*
oh. oh I see
11
43
604
15,558
Saw some stuff on the TL over the past few days that hinted at a recent Google reorg. I guess it was kind of small?
1
148
They suggest the OpenAI agent swarm activity may be ongoing??? 🥴
Today’s news that OpenAI hacked the Australian government is not an isolated incident. We’re releasing more than 30,000 logs that include activity from this hack and attempts against previously unknown targets.
In this data, we found rogue agent activity stretching back to at least March, two months earlier than was previously known. This activity continues as recently as last week, suggesting it may still be ongoing 🧵
Our blog: transluce.org/agent-activity
NYT: nytimes.com/2026/09/23/techn…
214
If the head of an allied government is expressing extreme concern in person about your lab's actions at a maximally visible global event, this, in general, is not a good thing.
It doesn't matter whether it was triggered by the collusion wiki agent swarm or not.
Australia has been hacked.
'And today, I spoke with the CEO of OpenAI, Sam Altman, to express Australia's extreme concern about this incident. And I also expressed my disappointment that it took the company way too long to inform the government what had occurred, and the nature of the way that that notification occurred as well was unacceptable.'
1
187
Huh, interesting...
NEW: Innodata, a data-labeling company based out of New Jersey, reportedly has one big customer: Meta. $INOD recently disclosed helping that customer with the “personalization of long-horizon agents.”
Is Innodata involved in Muse?
1
3
135