AGI forecaster. In 2022 I predicted an AGI timeline of 2027 MIT dropout
Joined February 2021
- Tweets421
- Following228
- Followers2.6K
- Likes1.2K
mei retweeted
There’s a huge blindspot in the inference market today: it’s built almost entirely for coding agents.
Voice is how billions of people will interface with AGI. That’s what we’re building at @hopper_research : the inference cloud for voice
Excited to launch today at withhopper.com
mei retweeted
Cancer patients with the same diagnosis often have very different tumor biology.
Molecular testing can reveal underlying differences, but that's slow and expensive.
Introducing Holon-S1, our platform to predict genes and biomarkers from routine pathology slides.
• beats two other models at predicting spatial gene activity in a controlled comparison
• has identified potentially novel biomarkers in breast cancer and colon cancer
• evaluated across 16 organ types and more than 15 scanners, using slides from both formalin-fixed, paraffin-embedded and fresh-frozen tissue.
We believe Holon-S1 has the potential to help researchers find new biomarkers in slides they already have, discover treatment targets, and bring more personalized cancer care within reach!
mei retweeted
Introducing Omnii-1, our most powerful genome language model, and a major step toward general biological intelligence
In a world first, Omnii-1 designed the first AI-generated bacteriophages to be administered to a human patient with an antibiotic-resistant infection under compassionate use
mei retweeted
Today, Dream Labs is coming out of stealth with Dream-1, a robot foundation model that controls bimanual robots it has never seen, on tasks it was never shown, in places it has never been, from $3/hour.
Dream API is live for everyone, today. 🧵
mei retweeted
was at a community event and was asked a somewhat sad question: should developers be afraid of the labs eating the new possibilities jev opens up?
with AI (system one) + software (system two):
1. there's no going back to a world without system one models
2. you control the software!
3. there's no way AI could beat the efficiency of this combo
4. you don't even send the model providers your whole program logic
I think the days of "wrapper" apps will be seen as some of the darkest in software because people essentially did all of their programming in a system message and sent it over to a third party who was incentivized to bake that behavior into their 1st party product...
keeping your logic in code and only asking AI for the minimum reverses that and keeps all of your logic yours. if you decompose enough, model providers can't even figure out what you're working on
AI founders: be aware that you have until the next election to get your startup acquired / acquihired. The previous Democrat FTC blocked nearly all tech acquisitions
You either build AGI or die trying. Or you get acquired by nvidia before the Democrats/FTC block acquisitions
mei retweeted
Excited to finally share that @phinity_ai is now out of stealth! We're announcing our seed fundraise and that we’ve reached 8-figures in annualized run rate.
We’re already working with some of the largest foundation model labs. Our mission is to build fully autonomous chip design so AI can help us search beyond the limits of human engineering and discover computational systems we would never find on our own.
We’re fortunate to have the support of @uncorkcap, which led our $5.2M seed round, with participation from @Moxxie including @JeffDean, following our pre-seed led by @pearvc!
It’s been an amazing few months, and I couldn’t be more grateful for the opportunity to build the future of chip design with such a talented team.
You can read more about what we’re building and why in our full announcement. phinity.ai/blog/introducing-…
We’re hiring across all disciplines - if you’re interested in working with us, please reach out to me or @aadi_nash.
mei retweeted
launched a model
accidentally served trillions of tokens a day
29.4% of the fortune 500 showed up
raised a really big series A from @a16z
it's been 3 weeks
we would like to sleep now
mei retweeted
New: Decision Model Rankings!
See the spend share and token share of different decision models and task types, including Relevance, Correctness, Instruction Following and more.
@typesafeai is leading all categories today.
mei retweeted
Jev launched on a Thursday. By Sunday, the @TypeSafeAI team had served a trillion tokens with no signs of slowing down.
Learn how they scaled Jev on Modal.
mei retweeted
Announcing $38M Seed & Series A from Sequoia, First Round, and Hummingbird.
Healthleap's mission is to improve the health outcomes of as many people as possible.
Results so far:
• faster healing resulting in 1.6 shorter hospital stays per patient flagged as at risk by HealthLeap (Penn)
• 39% more appropriate diagnoses, same staff (Cedars-Sinai)
• 3 → 50+ hospitals in the past year
We're hiring across all roles.
mei retweeted
FLUX 3 is the strongest world model in the world.
FLUX 3 is #1 on Physics-IQ, @GoogleDeepMind's benchmark for physical world understanding in video models.
How Physics-IQ works: given the start of a real filmed experiment, can a model predict what physically happens next? Tests span fluid dynamics, optics, solid mechanics, magnetism and thermodynamics, covering how liquids flow, light behaves and objects interact.
In most cases, we substantially outperform all other frontier models, Seedance 2.5, MiniMax H3, Gemini Omni 1.1 Flash, Veo 3.1 and Sora 2, Cosmos3.
The FLUX 3 base model outperforms all other foundation models, while FLUX 3 paired with a physics verification layer scores even higher.
The FLUX 3 backbone, with its strong physics understanding, gives FLUX 3 Action it’s strong action prediction for robotics, while also giving FLUX 3 Video physically accurate motion when you’re creating content, making ads, films and product shots.
mei retweeted
Today we're launching WebCloneBench: a benchmark which measures a model's ability to explore, copy, build and test various web apps.
For each task, we give the model a reference site from one of Gmail, Google Sheets, Jira, DocuSign, Shopify Admin, Craigslist, Figma, Slack and Zendesk. It is instructed to build a fully feature-complete clone of every part of the application with an 8 hour time limit.
mei retweeted
Took a while, but here it is, a mega write-up on language models for text classification and, of course, Jev.
It's basically a visual guide to RNNs, CNNs, transformers, and calibration, with hands-on experiments on accuracy and efficiency.
mei retweeted
We gave Jev 2,029 real phone calls.
No transcripts or audio; it never heard a word. Our AI receptionist's calls were reduced to pure structure, meaning turns, tool calls, workflow stages and timing.
During the calls, Jev made 38,012 turn-level forecasts at 118 ms median latency, reviewed every call with five typed questions and produced 10,145 answers in 26 seconds with 256 requests in flight.
The experiment was zero-shot, with no fine-tuning or examples from our data. We compared Jev's forecasts with what actually happened in the EHR.
By the halfway point, Jev could meaningfully separate calls that would book from those that wouldn't (AUC 0.78), and near the end it ranked them correctly 94% of the time.
Even though Jev over-focused on visible errors our agent usually overcomes, it's still pretty incredible that it analyzed thousands of real calls in seconds for only $3.