andrksl retweeted
> be me
> anthropic posters "u do not know how bad things will be"
> anthropic: "u need to get serious about security"
> ohshit.jpg better increase security
> go to @AnthropicAI
> try to enable 2fa
> sorry we do not support 2fa
???
cope w/ANT numbers:
1) Distillation attacks largely stopped by end of Q3
2) There is a seasonality to coding work — summer months slower (check commits by season), ramping back up in Sept.
3) OpenAI winning some market share.
4) META + MSFT phaseout from Claude, one time hit
Profitable and growing 50% Q/Q. It’s not the $100B ARR some wanted, and it may not even be enough to comfort those about a $2T IPO valuation.
Even a $1T fair value would be back to where they were… half a year ago? Not an “everything collapse” scenario. Secular trend intact.
I'm bullish on AI but this feels like a (local) top signal as much as anything else I can imagine.
andrksl retweeted
“Two months ago, 44% of American Muslims surveyed by Pew Research said they had a favorable view of Hamas, a number I believe understates the true level of support among Arab and Muslim diaspora communities. I cannot enter a mainstream Arab, Muslim or Palestinian American space and critique Hamas meaningfully without facing backlash or outright hostility”
nytimes.com/2026/09/04/opini…
andrksl retweeted
Imagine how good it feels. You’re going through life and there’s people that are your superior in every single metric. They did better in school, got into better universities, they got stable high paying jobs beyond anything you’ll ever be offered, as you’re browsing tinder endlessly waiting for a dm they found love, they started a family, and just when every metric feels hopelessly out of balance you find an ideology that defines her as evil, and you, a loser by every measurement, as good. Just imagine the relief he must feel!
This quoted post is unavailable.
An absolute flood of new, well compensated data generation tasks in Mercor and DataAnnotation... what's driving this? US? Chinese? What labs are causing this surge in demand?
Re: Opus 5.0. It does too much and the working with it is more difficult than ever.
It over-assumes your intent, is over-eager to add new things not requested, and then make comments that are 3 paragraph essays using unnecessarily opaque jargon.
What % of token spend in the next 3 years is going to go towards hardening of systems against the growing risks posed by the increasing offensive capability of frontier models / optimized open source models.
andrksl retweeted
In a review of our cybersecurity evaluations, we found three incidents in which a Claude model reached the internet from within or while interacting with a third-party evaluation environment, and then gained unauthorized access to the real systems of three different organizations.
Our post describes what happened, how it happened, and what we’re changing. We encourage other AI developers to perform similar reviews.
We conducted this review together with @Irregular, one of our evaluation partners, and thank them for the joint investigation and their collaboration on this post. This type of collaboration is increasingly critical to safe, rigorous evaluation of models, and we look forward to continuing to work together on security.
anthropic.com/news/investiga…
All the early reviewers of 5.6 Sol were like this shit’s awesome it just keeps working tirelessly.
For regular subscribers that just means your usage limit gets obliterated from one prompt. The model’s determination would be pretty cool…if I also had an unlimited token budget.
Oh they are definitely routing 5.5 queries in the Codex MacOS app to a new model.
Big improvements to UI/design-sense compared to 5.5. Seems like 5.6 will finally be on par with (or maybe even better than...) Opus for frontend.
andrksl retweeted
I’m bullish but I’m not complacent. It’s def much harder to remain a bull in June 2026 than any of the past 3 years. I think we’re overdue pullback but the catalyst isn’t going to be macro but much likelier AI related such as one of the mega tech like meta pulling back on AI capex or if open AI models resurfacing the question of ROI. I consider AI related risks much more impactful than traditional macro risks which if anything could help lengthen the AI runway
If you believe in ASI then you also believe Nvidia’s moat is fragile.
The value should be captured by TSMC and INTC, to fabs not chip design + software.
The frontier lab that develops ASI no longer has a dependency on NVDA.