Ali Emadi retweeted
We're giving away a Hogwarts Legacy custom wrapped NVIDIA GeForce RTX 5090 to celebrate its official release with DLSS 5 🪄
To enter:
🟢 Share this post
🟢 Comment #RTXON
Terms and Conditions: nvidia.com/en-us/geforce/con…
Here is an example of how you don’t need to be a lawyer to have a brilliant legal mind…
hear me out
there are four NVIDIA H100s in every Waymo
just call a Waymo and keep changing the destination continuously for 5 years so it never ends the ride
you now legally own the car and $200k worth of GPUs
do this with 5 cars and you're a millionaire
1
16
Ali Emadi retweeted
Databricks CEO Ali Ghodsi says if frontier models stopped advancing, it wouldn't matter for most orgs because enterprise AI is bottlenecked on context, not intelligence:
"There's no agentic work that's automated the whole enterprise. That has just not happened. Why is that? The real reason is the models are smart enough, but they just don't have the context that exists inside of any organization."
"They have not been in every meeting. They don't know what's in everybody's heads. They don't know all the processes. There's always a couple of employees who know everything in every organization. You go tap on their shoulder, and everybody's like, 'Oh my God, what would happen if he or she quits?'"
"If you just fused that and gave that context into the AI models, just the frontier today, I think there's so much productivity gains you could get for any organization on the planet."
"For that, we actually don't need smarter models... If the frontier doesn't advance, it doesn't actually matter for the vast majority of organizations on the planet. They're just so far behind in the adoption curve of actually automating things and getting value out of this stuff."
@alighodsi
Read the latest Profiles of the Future by @bhorowitz and @mikemcg0 on Databricks & the Future of the Firm: a16z.news/p/databricks-and-t…
Ali Emadi retweeted
This morning, we joined a panel of government and industry experts at the Billington Cybersecurity Summit to discuss the importance of Software Bill of Materials (SBOM) to supply chain security. Explore our SBOM resources: go.dhs.gov/5uc
ALT A panel of five individuals seated at a long table draped in a blue cloth, each with a nameplate. Bottled water and microphones are placed in front of them. One person is speaking, while the others listen attentively.
Ali Emadi retweeted
AI Hallucinations Are Exploding In U.S. Courts, New Study Finds zerohedge.com/markets/ai-hal…
We no longer need anecdotes to cut through the sales puffery of some AI startups (and their VCs) claiming that the UI they built on top of someone else’s intelligence produces better results.
Independent research found that general-purpose LLMs outperformed specialized clinical AI tools on all three medical evaluations tested.
nature.com/articles/s41591-0…
63
Who is going to tell them that LLMs are not deterministic?
This Startup’s AI Found Critical Vulnerabilities That Anthropic’s Mythos Missed
Startup Depthfirst claims its AI found some major flaws in tools that help run much of the internet, all for a tenth of the cost of Anthropic’s comparable model Mythos.
forbes.com/sites/thomasbrews…
37
This is an excellent Friday read…
We started by investigating why Claude chose to blackmail. We believe the original source of the behavior was internet text that portrays AI as evil and interested in self-preservation.
Our post-training at the time wasn’t making it worse—but it also wasn’t making it better.
39
Ali Emadi retweeted
LIVE: Looking for more space action following Artemis II? We have you covered!
Tune in to watch a cargo spacecraft arrive at the @Space_Station, delivering over 11,000 pounds of science, supplies, and equipment for the crew in orbit. xitter.cf/i/broadcasts/1rGmqolPL…
Great read.
Before AI minted a new class of billionaires, before OpenAI and Anthropic existed, there was a race to acquire DeepMind on.wsj.com/4rT7tB3
102
This graph is the latest lightning rod in the “AI is coming for jobs” debate.
And the comments show how most people are missing the point entirely.
Yes — it shows where AI agents are hitting hardest.
But comments and posts show almost all are looking at it the wrong way.
Stop thinking in jobs.
Start thinking in tasks.
The wrong question:
“Which jobs will AI replace?”
The right question:
“Which tasks will AI replace?”
AI doesn’t eliminate professions, unless that profession is just a task.
Software engineering isn’t going anywhere.
It’s leveling up and stronger than ever.
But manual coding?
That’s already changed forever.
Software engineering makes up ~50% of agentic tool calls on our API, but we see emerging use in other industries.
As the frontier of risk and autonomy expands, post-deployment monitoring becomes essential. We encourage other model developers to extend this research.
2
3
311
San Francisco is back.
People used to say San Francisco was dead.
Okta CEO @toddmckinnon says those fears were totally overblown:
“How many people who moved to Miami are now coming back?”
“Is the government perfect? No. Is any government perfect? No. But it's an amazing place, and all the people still want to come here.”
1
1
137
Thank you, Mayor Mahan (@MattMahanSJ). You’re standing on the right side of history.
California is home to the largest population of Iranians outside of Iran. I am proud to stand with them as they fight for freedom in the face of brutal, senseless repression. As Governor, I can’t set foreign policy — but I will stand up for human rights and work to ensure that our state lives up to the ideals that inspired so many Iranian-Americans to settle here in the first place.
VMware by Broadcom has been amongst the top 3 contributors to the Kubernetes code base both in the past 10 years and in the past 5 years. 🎤🤜🫳
k8s.devstats.cncf.io/d/9/com…
2
105
Demis (@demishassabis) isn’t being a “cautious optimist” here—he’s being a balanced optimist. There should be no negative connotation to what he’s saying.
He’s calling for basic standards and highlighting that they need to be globally adopted to work.
Standards don’t slow progress; they enable it—and proactive balance now reduces the risk of the backlash and overregulation that follow preventable failures.
On the path to AGI, @DemisHassabis is a cautious optimist. 🔅
In an interview at @WEF, he explains why AI development requires a balance of ambition and responsibility – emphasizing that it could be one of the most beneficial technologies for humanity. → weforum.org/meetings/world-e…
1
112