I love that AI companies are all like "we just have a marketing problem" when regular people are having AI experiences like this all over the place every single day

Bill Grueskin@bgrueskin.bsky.social · yesterday

Gotta love the “or stocks” part. Like Gemini realized its first answer was nuts so it tried to wedge in a better one.

🧵 One thing that concerns me about the current AI mania is what I see as people embracing full on nihilism - the "nothing matters" philosophy. - "LLMs have a massive environmental cost? Doesn't matter, because *future smarter* AI will solve climate change." 1/7

Fascinating experiment: current AI systems lack creativity to reliably pursue research arxiv.org/abs/2607.27191 - poor judgment about the bar for publishable research - uncreative responses in research design - ineffective backtracking from dead ends - poor resource awareness - instruction drift

Can AI agents conduct open-ended AI research? Early evidence from two case studies

Forecasts of explosive AI progress hinge on AI agents automating AI research. But evidence on whether agents can carry out open-ended AI research is thin. Current evaluations either test agents on nar...

arxiv.org

Don't be a meat proxy gruhn.me/blog/2026-08...

No one: Not a soul: Nice try, but not even them: Sam Altman: What if we replace everyday conversations between parents and children, the very threads that tie a family together, with synthetic podcasts?

New feature in AgentScan GitHub app: the honeypot comment. This comment is intended to welcome regular users, but it also contains a hidden comment that only AI agents will read. This will prompt them to write a specific custom string that will immediately close the PR and add a label as automations

As of today, the use of AI becomes more transparent in the EU. Here’s how EU law makes sure AI works for you, in full openness ↓

🔗 From today, EU regulation around labeling AI generated content goes into effect. The Commission provides icons that can be used digital-strategy.ec.europa.eu/en/policies/...

When will people learn that OpenAI and Anthropic have no problem lying in order to get more hundreds of billions of investment, just so that they can set that money on fire in their hopelessly unprofitable business? Don't fall for it. You look credulous if, at this point, you believe these people.

Honza Dvorsky@czechboy0.dev · 4d ago

Bad news, once again you've been had by an AI company's marketing department. At least one of their "proofs" is wrong. I recommend being highly skeptical of product marketing materials from desperate, unprofitable companies built on theft and hype. philarchive.org/rec/NIEWTC

If I’m keeping score correctly, so far the AI revolution has given me: - considerably worse software than before - considerably more expensive devices to run that software - considerably less opportunity to make the money needed to buy those devices

Anthropic's Claude is powered by racism. That's who Musk is running these DCs for. In exchange for billions of dollars.

Hypervisible @hypervisible.blacksky.app · 5d ago

The fact that it’s poisoning Black people is not seen as a negative to these folks.

None of this shit is inevitable. No part of "AI", not even LLMs, are inevitable. If enough people push back, these companies have no choice but to get rid of these harmful products just as quickly as they introduce them.

Joseph Cox@josephcox.bsky.social · 5d ago

Update: Google just sent me a statement saying it is “rolling back” this AI feature that let anyone make completely bullshit satellite images. Comes after our demos of the tool and those from OSINT researchers. One of the worst AI tools I’ve ever seen bsky.app/profile/jose...

My car went rogue after I turned on cruise control and jumped out

Timnit Gebru@timnitgebru.blacksky.app · 5d ago

We're in the era of incompetence & cybercrimes headlined as "unprecedented model capabilities gone rogue." So OpenAI & Anthropic are trying to one up each other with such incompetence because the "press release as a service" performing media & clueless politicians parrot pre IPO CEO talking points.

ANTHROPIC: Our son loves humanity and for the low price of trillions and trillions of dollars will usher in an era of unprecedented peace and prosperity! OPENAI: Our horrible oafish son is hacking web sites without our permission. ANTHROPIC: (eyes narrowing) Our son is also evil

Hackers in America threatened with conspiracy charges just for researching company’s computer security and AI companies get to turn actually hacking other corporations into marketing material.

Brian Honan@brianhonan.bsky.social · 6d ago

When I think of all the people in the hacking community who faced the full brunt of the law, civil and criminal, and then see AI companies go "oopsie our bad" when causing damage to innocent companies it makes me angry and frustrated.

It’s really starting to grind me down

If pocket calculators regularly and randomly gave you incorrect answers to your queries like ChatGPT does, we would have left the technology behind, not insisted that 2+2=5 is HERE TO STAY, BABY.

Uncle Fuzzzy 🦆@unclefuzzzy.bsky.social · 4mo ago

Funny, that’s what we thought about pocket calculators in 1974

Spreading this dangerously unreliable technology is now leading to human rights violations. When will this madness end? www.theguardian.com/uk-news/2026...

Home Office used ‘AI hallucinated’ information to refuse asylum claim, judge suggests

Upper tribunal unable to find document cited in case and claims it is ‘analogous to relying on bogus evidence’

theguardian.com

finally got a link to share with people who try and say "oh you DON'T LIKE AI??? what about ALPHAFOLD!!!!" as if it has anything to do with any of this AI capex

Engadget@engadget.com · last wk.

DeepMind's award-winning AlphaFold team is no more.

I'm baffled by AI boosters genuinely standing by this guy. Really? This is your leader? Yikes. History won't judge this stupid era and the people who caused it well.

Dare Obasanjo@carnage4life.bsky.social · last wk.

“What we need to explain to the world: Number one, we are close to creating a genie that can grant any wish. Number two, we are going to make sure that our first wishes broadly benefit humanity” - Sam Altman, CEO of OpenAI — This is definitely the opposite of Anthropic’s AI kills all jobs pitch.

There is something so compelling and beautiful about robot fail videos

When you're excited about a new project but then you see this

Dario and Boris are not telling the truth in interviews? WAT what a completely unexpected revelation that nobody saw coming

Gergely Orosz@gergely.pragmaticengineer.com · last wk.

So many interesting details. Here's one: "One thing I’ve heard from some people is “we have two humans and a bunch of agents.” I reply that this isn’t where we’re at: we still have teams with 6-8 people on average." - Katelyn Lesse, Head of Engineering, Claude Platform, Anthropic

I really love OP’s implication that character-level manipulation is something LLMs are intrinsically unsuited for by nature of how they work, but almost everything else they are supposedly capable of doing (reasoning, returning factual data, holding a conversation) isn’t.

Do you ever think about how Sam Altman has never, not once in his life, run a profitable business? Because I think about that all the time. THIS is who we're allowing to drag the global economy to the brink of collapse.

Al Jazeera English@aljazeera.com · last wk.

The 'singularity' refers to the point at which AI outpaces human intelligence and becomes difficult to control.

On controlling AI: If your "strong constraints and guardrails" come in the form of context and prompts, you have no constraints or guardrails. What you have are hopes and prayers. LLMs don't follow instructions; they regenerate and recombine contextually appropriate tokens.

The AI industry is truly fucked. This is historic level of bipartisan pushback. And it's only deserved. From the trillionaires driving this at the top to the AI-boosting techfluencers at the bottom, I can't wait for them all to go away and stop bothering the rest of us. Let's pop that bubble 🫧📌

404 Media@404media.co · last wk.

Police arrested a teacher for clapping during a city council meeting last week. They went to the meeting of Emporia, Kansas City Commission to speak out against a proposed hyperscale data center that would sit on 1,000 acres of rural land. @mjgault.bsky.social reports www.404media.co/person-oppos...