Lightnews — Scholar-powered news

Chris Paxton @cpaxton.bsky.social · 2h

Agility is looking for ai researchers who want to work on the frontier of whole body control + imitation learning for humanoid robots: www.agilityrobotics.com/about/job-po...

Agility Robotics - Job Posting

Join the Agility team and help build the humanoid robotics industry.

www.agilityrobotics.com

2 7

Reposted by Chris Paxton

Ethan Mollick @emollick.bsky.social · 5h

One of the first randomized controlled trials testing whether GenAI boosts revenue, not just productivity.

It does.

A large, mature international ecommerce platform, using older GenAI tools found most of them, from customer service to marketing workflows, led to large and significant revenue gains

5 7 36

Reposted by Chris Paxton

spleenly @spleenly.bsky.social · 11h

Avoiding the whole AI video phenomenon by never watching any video ever.

14 36

Chris Paxton @cpaxton.bsky.social · 8h

A new dataset of humanoid robots performing whole body actions humanoideveryday.github.io

1 9

Reposted by Chris Paxton

Ryan Moulton @moultano.bsky.social · 18h

This stuff is so sad. It should not be up to a private organization to make sure we don't have lead in our food.

paris martineau @paris.nyc · 1d

my latest investigation for @consumerreports.org is based on months of reporting and 60+ lab tests of leading protein supplements

we found that most protein powders and shakes have more lead in one serving than our experts say is safe to have in a day (🧵)

www.consumerreports.org/lead/protein...

Protein Powders and Shakes Contain High Levels of Lead - Consumer Reports

CR tests of 23 popular protein powders and shakes found that most contain high levels of lead.

www.consumerreports.org

6 13 99

Reposted by Chris Paxton

Ethan Mollick @emollick.bsky.social · 16h

Every few months on BlueSky there is a hot story of how AI is going to go away soon: model collapse, walls to AI development, legal issues, now financial bubbles.

But even if there was a bubble that popped, AI would not go away: 10% of humans use it weekly, increasing use cases in companies, etc.

14 12 130

Reposted by Chris Paxton

Ryan Moulton @moultano.bsky.social · 1d

Someone should start a nostalgia-bait account that's just pictures of the present day, college kids playing volley ball on the beach in Santa Cruz, leafy New England villages, cub scouts roasting marshmallows around a campfire. It's all still happening, you just aren't doing it.

3 7 68

Reposted by Chris Paxton

Ted Underwood @tedunderwood.com · 1d

I have a seared-in memory of following my dad around at a conference when I was ~12, and asking whether someone was right — and his response that “what matters is not so much having the right answers but finding people who are identifying good questions.” Felt the world flip upside down.

brennan @brennan.computer · 1d

the way I learned it is being in the same room with other people who ask extremely good questions

(and then trying to increase the number of rooms I was in with such people)

1 2 44

Chris Paxton @cpaxton.bsky.social · 2d

It's fascinating how competitive the open source scene has become in China. i've never even heard of Z.ai

Kevin Schaul @kevinschaul.bsky.social · 2d

New from me: Last year, the best open-weight AI models were made in the U.S. Now, they are all made in China.

More data and what it means -> 🎁 wapo.st/4nPUBud

Chart titled Chinese companies make the most popular free AI models

2 1 13

Reposted by Chris Paxton

Kevin Schaul @kevinschaul.bsky.social · 2d

New from me: Last year, the best open-weight AI models were made in the U.S. Now, they are all made in China.

More data and what it means -> 🎁 wapo.st/4nPUBud

10 42

Reposted by Chris Paxton

Ethan Mollick @emollick.bsky.social · 2d

This relatively short essay by Jack Clark (from OpenAI & Anthropic) is a good indicator of the attitude of many people inside the AI labs, and what they think is happening right now in AI.

You do not have to believe him, of course, but it is worth noting: importai.substack.com/p/import-ai-...

8 12 76

Reposted by Chris Paxton

Phillip Isola @phillipisola.bsky.social · 4d

Over the past year, my lab has been working on fleshing out theory + applications of the Platonic Representation Hypothesis.

Today I want to share two new works on this topic:

Eliciting higher alignment: arxiv.org/abs/2510.02425
Unpaired learning of unified reps: arxiv.org/abs/2510.08492

1/9

1 30 130

Reposted by Chris Paxton

Tim Kellogg @timkellogg.me · 2d

this is *wild*

a series of papers indicating that text-only LLMs still have an idea of how audio or visual modes work

Phillip Isola @phillipisola.bsky.social · 4d

Over the past year, my lab has been working on fleshing out theory + applications of the Platonic Representation Hypothesis.

Today I want to share two new works on this topic:

Eliciting higher alignment: arxiv.org/abs/2510.02425
Unpaired learning of unified reps: arxiv.org/abs/2510.08492

1/9

5 2 49

Chris Paxton @cpaxton.bsky.social · 2d

What's the practical difference between the approach at the two companies? Is it really just data, or is it something deeper?

1 1

Reposted by Chris Paxton

Nathan Lambert @natolambert.bsky.social · 3d

Recurring frontier lab gossip:

OpenAI has best post-training/rl and has pushed it super hard on weaker pretraining.

Gemini has spectacular pretraining. Making a reasoning model was super easy for them & OpenAI folks were surprised

Anthropic? Secretive i guess.

5 2 75

Reposted by Chris Paxton

Tim Kellogg @timkellogg.me · 2d

if this is true, i think you should expect Gemini to soon dominate

we keep getting more and more confirmation that reasoning begins in pre-training

today’s evidence: arxiv.org/abs/2510.07364

maybe Gemini 3 is the tidal shift where Google gains a permanent lead

3 3 43

Chris Paxton @cpaxton.bsky.social · 5d

Million? Those are rookie numbers

3

Chris Paxton @cpaxton.bsky.social · 5d

Because they're hard in boring ways, highly regulated, and low margin

2

Chris Paxton @cpaxton.bsky.social · 5d

I need to be more ai first

Tim Kellogg @timkellogg.me · 5d

tiny AI-first teams are drawing billions

the state of AI report is out and one of the findings is that AI is enabling small companies to do huge things

www.stateof.ai

Introduction | Research | Industry | Politics | Safety | Survey | Predictions
AIR STREET CAPITAL | 98

⸻

AI-first companies are now generating tens of billions of revenue per year

A leading cohort of 16 AI-first companies are now generating $18.5B of annualized revenue as of Aug ‘25 (left). Meanwhile, an a16z dataset suggests that the median enterprise and consumer AI apps now reach more than $2M ARR and $4M ARR in year one, respectively. Note that this will feature significant sample bias, as evidenced by the bottom quartile not being close to $0. Furthermore, the Lean AI Leaderboard of 44 AI-first companies with more than $5M ARR, <50 FTE, and under five years old (e.g. includes Midjourney, Surge, Cursor, Mercor, Lovable, etc) sums over $4B revenue with an average of >$2.5M revenue/employee and 22 employees/co.

⸻

Chart legend (stacked area chart):
• OpenAI (red)
• Anthropic (blue)
• Anysphere (Cursor) (green)
• xAI (gray)
• 14 Others* (dark gray)

*Includes AI native apps with >$50M in annualized revenue: Midjourney, Perplexity, Abridge, Synthesia, Replit, EliseAI, Lovable, Glean, ElevenLabs, Cognition (incl. Windsurf), Runway, Cohere, Jasper, Harvey
Source: The Information reporting

⸻

Bar chart (Consumer vs. Enterprise)

Quartile Consumer Enterprise
Bottom quartile $2.9M $1.2M
Median $4.2M $2.1M
Top quartile $8.7M $5.3M

⸻

stateof.ai 2025
Bottom left logos: T and a16z

1 1 9

Reposted by Chris Paxton

Tim Kellogg @timkellogg.me · 5d

tiny AI-first teams are drawing billions

the state of AI report is out and one of the findings is that AI is enabling small companies to do huge things

www.stateof.ai

2 2 32

Chris Paxton @cpaxton.bsky.social · 5d

Figure 03 is impressive and continues the company's steady progress towards general purpose humanoids. A few things I am skeptical of or would like to see more: navigation and whole body control are behind SOTA, environments seem too canned, but great.

open.substack.com/pub/itcanthi...

1 2 11

Reposted by Chris Paxton

Matthew Yglesias @mattyglesias.bsky.social · 6d

All the terrible stuff we were worried about that mostly didn't happen in Trump's first term is ... actually happening

www.slowboring.com/p/the-author...

The authoritarian menace has arrived

And so has the backlash — but the resistance needs to be smarter

www.slowboring.com

5 23 130

Reposted by Chris Paxton

Matthew Yglesias @mattyglesias.bsky.social · 5d

The Orange Man is in fact very bad.

28 47 520

Chris Paxton @cpaxton.bsky.social · 5d

I genuinely have never understood the appeal of any of these. There are no computers out there. What would i do, it would be so boring

SE Gyges @segyges.bsky.social · 6d

your software engineering career depends on not developing a fascination with opening a farm, carpentry, or living in a camper until you have at least 8 years of experience. if you get the bug early it's joever for you

2 17

Reposted by Chris Paxton

Alexander Doria @dorialexander.bsky.social · 6d

Unfortunately realizing a bit late the recipe to get 2 billion was to not publish on HF.

1 1 11