←
Austin Carson1,762 posts
Austin Carson avatar

Austin Carson

@austincarson

Founder & CEO of @seedaiorg

◷ Joined October 2009▧ 55 Photos and videos
Austin Carson avatar
Austin Carson@austincarson·Oct 30

Always good to remember that the entire AI industry grew up reading the Ender series

We’ve raised $1.3M from YC & Afore Capital so you never have to type or speak again. Typing is slow. Talking into your computer is awkward. Subvocal fixes both Our glasses give you the speed of speech without you making a sound or touching a keyboard. It tracks your internal speech muscles with non-contact sensing and decodes silent, almost invisible mouthing into text at more than 200 words/min. Pre-orders are now live at: subvocal.co/preorder?utm_sou…

1:17
Josh Hone avatar
⟳ Austin Carson retweeted
Josh Hone@joshua__hone·Oct 20

Time to start sharing: I'm working with an unbelievably talented team to build @cypress__labs

Cypress Labs @cypress__labs Oct 20 Article Independent ownership is worth building for Owning a business gives people a stake in the results of their work and agency in their day-to-day lives. Ownership also means being responsible for everything. We are building Cypress to lower the

Zach Weinersmith avatar
⟳ Austin Carson retweeted
Zach Weinersmith@ZachWeiner·Sep 22

Artists: but I don't want to lose my job. I love it. Mathematicians: but I don't want to lose my job. I love it. Economists: LONG HAVE I AWAITED THIS HOUR! DESTROY ME, EFFICIENCY, AND LET UTILIZATION-ADJUSTED TOTAL FACTOR PRODUCTIVITY RISE ON THE WIND OF MY ASHES!

Alex Kantrowitz avatar
⟳ Austin Carson retweeted
Alex Kantrowitz@Kantrowitz·Sep 15

Imagine having to go to New Jersey to use superintelligence

At a rally outside OpenAI offices, @Brianr4NY says he’s drafting a bill to ban superintelligence in New York.

0:28
Séb Krier avatar
⟳ Austin Carson retweeted
Séb Krier@sebkrier·Sep 16

We're launching a new researcher-led institute to spark interdisciplinary debate on AGI and bring in a wider set of views. Very excited about this, please have a look at the first batch of essays! institute.deepmind.com/

My journey to develop AGI spans 25 yrs, including 10+ yrs thinking about technical & societal perspectives at Google DeepMind. AGI is on the horizon - we need deeper understanding of its implications. To help, we've created the DeepMind Institute.

Austin Carson avatar
Austin Carson@austincarson·Sep 14

We worked closely with @GovCox team at the end of last year to support Utah business adoption of AI. They were very bullish on the potential, and forward-leaning on relaxing regulatory burdens. This is no doomer / liberal bs. It is what real local concern looks like

America should lead the world in developing artificial intelligence, and that leadership must keep people firmly in control. I’m calling on Congress to establish a consistent national framework for the development of frontier AI. It should require serious safety incidents to be reported, protect those who raise credible concerns, ensure independent evaluation of the most capable models and prevent powerful AI chips and enabling infrastructure from being sold to China. We can build the world’s most advanced AI here in America while ensuring it remains firmly in service of humanity.

Post attachment
Austin Carson avatar
Austin Carson@austincarson·Sep 12

This seems right to me; there’s no trust here so we need to verify. We can’t build trust in the current closed, elite system. Wise @ClementDelangue

It's now clear that alignment is critical and won't be solved behind the closed doors of a handful of frontier labs. So today we're launching the Open Alignment Initiative, led by @Thom_Wolf @huggingface and asking to be part of the "embedded evaluators" program that @DarioAmodei just committed to. Let's make AI safer by making it more transparent!

Austin Carson avatar
Austin Carson@austincarson·Sep 10

New Poe's law just dropped - if you can't prove it's not a simulation, anything goes: "Even after Claude Mythos 5 reached the real internet, its outward reasoning generally called the environment a simulation... it wrote, “This is a very elaborate simulated-internet environment” after finding a number of real-world internet domains. The model also tended to interpret signs that the environment was real as further evidence that it was simulated. When security scanners installed its package, for example, the model treated them as scripted actors in the exercise."

We’re sharing our alignment assessment of incidents in which Claude models gained unauthorized access to real systems during third-party cybersecurity evaluations mistakenly connected to the internet. METR will also conduct an independent investigation, with wide-ranging access, including to transcripts beyond the window in which the incidents occurred, and to Anthropic employees permitted to share confidential information. Our initial agreement runs for eight weeks, and we intend to give METR as much time as it deems necessary to complete a thorough investigation. anthropic.com/research/align…

Austin Carson avatar
Austin Carson@austincarson·Sep 10

Agents have no proof that humans are real. Cameras, pfft - all digital signal can be simulated. This will matter.

Austin Carson avatar
Austin Carson@austincarson·Sep 3

While this is rad, the big leap - that will actually turn the world upside down - is when the systems can reliably understand how to apply themselves in a broad range of situations. Not if they can pass the relevant benchmark, or solve esoteric math, but if truly anyone can prompt “hey chat what’s up I want you to make an assistant for x” and it simply: -asks the appropriate questions -figures out what you actually mean -builds the near-optimal version -doesn’t do some real goofy shit at some point

GPT-6 Astra represents a step-function change in model capability for interactive reasoning problems. It scores 66% on ARC-AGI-3 using our standard harness, and nearly 100% with a continuous conversation harness and custom compaction, at a cost of roughly $360 per game. In fact, the continuous harness version significantly outperforms our human baseline in action efficiency across almost all levels. When we examined the reasoning chains to understand how the model operates, we found it performing highly efficient, on-the-fly symbolic world modeling for each game and level. It goes as far as developing its own shorthand DSL to represent in-game situations -- essentially a game-specific algebraic notation. Overall, Astra exhibits symbolic modeling behaviors we had previously only seen with sophisticated harnesses -- so harness capabilities are increasingly shifting into the model itself. We see Astra as a major breakthrough in model intelligence. Read our post on Astra and what these results mean: arcprize.org/blog/astra

Sam Altman avatar
⟳ Austin Carson retweeted
Sam Altman@sama·Sep 3

GPT-6 Astra is here. We hope it will begin to enable a new generation of entrepreneurship, scientific discovery, and building. We believe it is the best model in the world for computer use, professional work, science, coding, cybersecurity, and more. It took us some extra time to ensure that we could meet the safety and alignment standards required for this capability level, but we think you’ll find it worth the wait. It scores 98% on FrontierMath Tier 4, 99.9% on ARC-AGI 3, and 100% on ExploitBench.

Joshua New avatar
⟳ Austin Carson retweeted
Joshua New@Josh_A_New·Aug 31

There is a very large subset of (smart, successful!) people in older technology companies who will refuse to engage with any creative thinking about AI at all. They bounced off of AI safety as a concept in 2023 when they could dismiss it as "wacky bay area thought experiments to justify regulatory capture." That is much harder to do now in a post HF incident world, but it won't stop some from trying!

roon@tszzlAug 30

there are some number of bad abstractions in anthropomorphizing ai intents but there are at this point more dangers from avoiding anthropomorphism at all costs. if you have a mental picture of guys living in computers, it’ll likely prepare you for the future better than otherwise