AI posts
People and companies shaping AI. Filter your watched accounts, catch up on recent posts, or find the most liked and viewed.
Compiled by the Toolstash team · Updated
209 watched accounts · 2977 matching posts · Showing up to 3,000 recent stored posts. Counts are snapshots, not live X totals. This page refreshes every minute while visible. Last collected .
@ValsAI ·
About half of U.S. adults now use AI chatbots, and questions about visas, work permits and immigration status are among the most consequential they ask. Can the models be trusted?
With the Talent Mobility Fund, a program of @RenPhilanthropy, we are releasing Immigration Bench to test whether frontier models are able to answer these high-stakes questions.
@paolino ·
Can't wait to have RCC in Berlin next year.
We'll have a special @berlinrb_org the day before!
@theo ·
I'm not gonna lie, this puts us in first place by default. Nobody has this working end to end yet.
@levelsio ·
At least 85 people this year died from Spanish train accidents, and likely it will grow to 100 at the end of this day
Very tragic and this makes Spanish trains one of the most dangerous modes of transport now
~100x more dangerous than flying commercial
~40x more dangerous than non-Spanish trains
And every trip is nopw as dangerous as jumping from a plane and skydiving
Two highly deadly train accidents in one year, that's so extreme and unlikely that it makes it at as dangerous as flying one of the most deadly airlines in the world, the Nepalese Yeti Airlines that keeps crashing due to the mountainous terrain
Pretty exceptional in a bad way
@theo ·
Day 28/
Julius cooked harder than I thought.
Multi-machine orchestration has been achieved internally.
One thread, multi-model, multi-harness, multi-app, multi-os, multi-machine, multi-network.
We're just getting started :)
@steipete ·
OpenClaw’s web UI now works offline and uses a smart websocket compression/reconnect logic.
Been flying to Vienna today and got plenty work done, despite spotty airplane wifi.
@jxnlco ·
Unironically mission aligned right
@mattpocockuk ·
Banger PR
Summary, evidence, merge danger on point
A great example of my /pr skill in action
https://github.com/mattpocock/course-video-manager/pull/2399
@petergostev ·
Astra was a great model but expensive, GPT-6.1-Sol was only slightly worse, but 80% cheaper, then GPT-6.1-Terra is slightly worse than Sol, but it is 50% cheaper.
So in <2 months, we got a model that it only somewhat worse than a model that is 10x more expensive. Good stuff.
@natolambert ·
I feel a need to defend the AAv5 index and open models at the same time. Benchmarking is messier, more opinionated, and trickier than most people give credit to.
@theo ·
Hopefully this makes more sense now lol
@GergelyOrosz ·
"When the going gets tough, the tough get going."
It's a saying that comes to my mind when I get close to giving up on something new/difficult/different
Found myself using it a lot more, recently. Just because my old routines no longer work as well, and I'm trying new ones
@theo ·
Day 27/
Run into a bug? Debug it in t3 code itself with cmd+k "triage"
This will: - Clone the T3 Code source for your exact version - Give context with a playbook to your agent of choice - Let you debug and fix the problems with your agent - Cut an issue if it's our fault
@DeryaTR_ ·
This new v5 of the Artificial Analysis Intelligence Index makes much more sense now. Some open-source models were apparently benchmaxxing and are not as close to frontier models from OpenAI, Anthropic, or the new Gemini Argon as they were in the previous version of the benchmark.
@theo ·
Day 26/
Interrupt compaction with "send now" if you change your mind
(also put the token # back in the auto-compact toggle, missed it too much)
@MistralAI ·
Voxtral Mini Transcribe 3 ranks #2 on Voice Arena’s Diarization + ASR leaderboard.
@steipete ·
Kinda funny how easy it is to spot bugs by agents vs humans. Someone at Disney still writes code by HAND?
@theo ·
Day 25/
"Boring Mode" sidebar is now available as an experimental option.
I know some people just can't give up grouping by project. Here you go. Enjoy.
@theo ·
I'm so mad that I have to release and maintain an os 🙃
@MiaAI_lab ·
my dgx sparks are getting reinforcements 👀
@MaziyarPanahi ·
it's crazy how cheap GPT-6.1 Sol is, given how strong it is!
it's one point behind GPT-6 Astra on Artificial Analysis, on general and on healthcare & medical tasks, for 4.5x less per task
i'm curious, what would you use GPT-6.1 Terra for?
@tszzl ·
experts suggest that risks of models modifying the cosmological constant are far-fetched and that we should be more concerned about current day risks like models modifying the fine-structure constant
@ElevenLabs ·
Aníbal, an engineer and professor, is losing his voice to ALS. With ElevenLabs, he’s recreated it so he can keep teaching.
We’re partnering with @AsociacionELA, Spain’s leading ALS association, to give people with ALS across Spain free access to our voice restoration technology.
@theo ·
To make up for this, I'm also making the Tailscale experience 10x better (arguably better than T3 Connect)
More soon!
@mattpocockuk ·
We used to talk about a "pit of success".
Humans would be dragged towards writing the right code by well-designed API's, a good test suite, sensible lint rules.
You couldn't make the pit too steep, otherwise the humans would get annoyed. There was a social contract.
Feels like what we need now is a black hole of success. An inescapable event horizon agents cannot cross.
Agents don't feel annoyed, so you can crank up the pressure to infinity. There is no social contract to break.
In other words, increase the forces on agents to the maximum. Make great code inevitable.
@theo ·
Bad news: t3os is actually really fkn good
@theo ·
Transparency here:
I don't think we're gonna make much money on this. Literally just trying to give people more tunnels without going bankrupt and/or standing up Yet Another Pile Of Payment Infrastructure™
This gives you more T3 Connect tunnels now. Don't overthink it.
@bentossell ·
i feel stuck with wtf i do for like 10 months
keep going round in circles. questioning and re-questioning everything. things feel exciting then dont.
what feels like a good business direction then doesnt.
touch grass, ai melting brains, i know i know but still
@MiaAI_lab ·
Maybe Artificial Analysis needs to do voxels.
@marclou ·
Congrats, you're the fastest-growing startup of the week on TrustMRR 👏
$4,884 MRR +2,253% growth MoM
This guy built a real-time geospatial intelligence platform. It shows flights, satellites, ships, and so many other things on a globe.
It's free, but you can get the Pro plan ($10,99/mo) to unlock cameras, alerts, and, again, so-many-other-things. It's like Palantir for B2C.
→ https://trustmrr.com/startup/godsviewai?metric=mrr&period=30d
