I’m curious what the next few months are gonna look like when Claude Cowork goes mainstream. We’re already getting peeks at this with Claude in Office, but not sure if most office workers are ready for the transition.

don't really want to argue about the ethics of genAI overall right at the moment but if you're giving OpenAI in particular your money please consider literally any other option

Running Claude Code locally with gpt-oss via ollama feels illegal docs.ollama.com/integrations...

The biggest lesson I’ve learned from brief stints studying/living abroad: It’s okay for the government to do good things. Some people won’t like it. That’s ok. It’s still our collective responsible to do good things. This lesson is almost impossible to learn in the U.S. today.

As someone who spent too much time debugging ROCm bc AMD barely makes it useable, I’m glad for this (if it works)

it’s unearned valor on sounding like AI when you grew up reading old books

it’s a great guide for product quality and some evals are better than no evals, but safety, security, and fairness should be more regularly included in eval harnesses www.anthropic.com/engineering/...

me, a social scientist, trying to continually convince software teams that they have social science measurement challenges, seeing this title 😱😱😱😱😱😱😱😱😱

despite the numerous anti-ai accounts on here, bluesky is still where I go to hide from the bad ai discourse on linkedin

A bridge builder doesn’t need to denounce every evil in the world to be moral, but they better say something about the guy who keeps building bridges that topple over

strange how laypeople think the “autonomous decision-making” of AI agents only applies to substantive and material real-world decisions and not like, deciding to make basic tool calls like querying a database

A @promptfoo.bsky.social + @credoai.bsky.social integration would solve so many AI auditability and risk measurement challenges. Is anyone building something similar?

even with the rise of fully-autonomous coding agents like Codex and Claude Code, it’s important to keep a close eye on them, case in point: CC almost committed my API keys in plaintext

The HTEs in this study track with observations from the professional world: those with strong baseline learning skills stand to benefit the most from AI-assisted pedagogy

I’ve personally encountered some truly atrocious CX with the latest generation of voice AI agents when calling customer service numbers. Many companies are shoe-horning these agents in to cut costs, but will probably see more customers churn toward competitors still relying on human agents.

introducing claude code to my 60+ yo tech-savvy father was similarly successful - he was able to set up his home automation repo and have cc write additional scripts

a direct observation of this phenomenon is the presence of “model welfare” in claude’s system cards - in fact, the entire structure of anthropic’s system cards supports this interpretation

I uh, made this. It was supposed to be a joke / concept-art thing that scrolls through the torrent of new AI/ML arXiv uploads too fast to read. But I think I iterated too much and made it almost usable.

@anthropic.com's recent (excellent) discussion on AI in Education entirely focuses on this frame: youtu.be/Uh98_aGhAuY?...

here's a hot for Bluesky take: a lot of tech-criticism is so bad and focused on a kind of stupid imagined STEM-humanities culture war that some companies have been able to build whole brands around being "evil tech bros 😈" making supervillain promises despite their shit being vaporware

Companies seemed to realize that "I want to talk to a human" is used all too often and they put up more barriers before the bot actually connects to customer support This is where I lose my patience with the whole thing, after the 3rd time the agent blocks me from doing it, as a paying customer

These kinds of incidents could be avoided if the people making public technology procurement decisions had really any knowledge of technology instead of uncritically relying on what tech salespeople tell them

i love giving entities that are more gullible than my grandma access to both my full home infrastructure and unfiltered external communication methods

From now on we are required by law to call them fair use machines

I've been trying to articulate why the fawning, complimentary responses from AI chatbots feel so insidious to me. I've finally figured out how to explain it. Wrote a long piece on how current model training and design choices threaten our critical thinking skills: maggieappleton.com/ai-enlighten...

Don't leave AI to the STEM folks. They are often far worse at getting AI to do stuff than those with a liberal arts or social science bent. LLMs are built from the vast corpus human expression, and knowing the history & obscure corners of human works lets you do far more with AI & get its limits.

🚨 New paper from us: Given they are trained on human data, can you use psychological techniques that work on humans to persuade AI? Yes! Applying Cialdini's principles for human influence more than doubles the chance of GPT-4o-mini agreeing to objectionable requests compared to controls.

Elon saying he’s going to reprogram his chatbot’s view of history is a perfect example of why these kinds of AI products aren’t the neutral “tools” people always defend them as. They’re little political projects reflecting the ideologies of their creators

Got distracted today & did a little experiment on ChatGPT 4o. I told it I had a question about AI & asked it to suggest six experts to contact. In one condition, I put "tough-minded and rigorous" before "experts". In the other condition, I put "friendly and kind" before "experts". ⬇️

Not good. Thank you to @wired.com for documenting and to the researchers for speaking up.

I feel like my ability to figure out whether someone knows their stuff when they're talking about AI has become so sharply honed the more time I spend switching contexts between technical and non-technical people

Anthropic’s MCP should be more widely used in enterprise

An X engineer posted this output from Grok to demonstrate how "good" their LLM is (CW: racism)

The left version of ‘run government like a business’ and ‘cut red tape’ is that there are a ton of asinine constraints that the public sector is under, things that would instantly be recognized as insane practices in a well run private firm, in the name of ‘small government’.

This is sort of the skeleton key to understanding a lot of shit: there 100% is a non-trivial amount of inefficiency and waste in the federal bureaucracy, and most of it is a *direct result* of ‘reforms’ that are meant to assuage the people who complain about “waste and inefficiency in govt”

The most effective responsible generative AI technique is to proactively gatekeep LLMs from decision-makers who don't know how transformers work

A lot of people don't know how useful Model Context Protocol is - easily Anthropic's most underrated feature. Allows Claude to use your computer and applications with natural language. Official and unofficial MCP servers, which are like extensions, here: www.mcpservers.ai

If you have to prompt engineer confidence scores, then you already know they're not going to be reliable. arxiv.org/abs/2412.14737

If you, like me, have a feeling of despair reading this news, I'd ask that you consider donating to the organizations who do this work: Global Fund to Fight AIDS, Tuberculosis, and Malaria: act.unfoundation.org/FJvB3vUCJUep... International Medical Corps: internationalmedicalcorps.org?form=Main

It's actually theft of taxpayers money when websites go dark. We paid for that data

okay so what legal authority does elon musk have to commandeer the office of personnel management and access reams of sensitive data? www.reuters.com/world/us/mus...

Today, we are publishing the first-ever International AI Safety Report, backed by 30 countries and the OECD, UN, and EU. It summarises the state of the science on AI capabilities and risks, and how to mitigate those risks. 🧵 Full Report: assets.publishing.service.gov.uk/media/679a0c... 1/21

positron is the best IDE for data science and it's been my daily driver for over a year

Another takeaway is that we've been Stockholm Syndrome'd to think pandas has a reasonable and intuitive API for data analysis

middle managers who've never written a single line of code or built an ml model before

Every time I visit the North End here in Boston, I'm grateful to not be swallowed up by a flood of liquid sugarcane extract

A strange disconnect I've been thinking about: in the 1940s, 2.5m Indian soldiers fought against fascism in WW2. Many of their children came to America in the 90s tech wave. Now some of their kids and grandkids support the same ideologies they fought against

#Boston friends: want to see how the city's services are measuring up? I built CityUpdate Boston - daily plain-language summaries of municipal performance data, from EMS response times to sign repairs. Check it out: hersh-gupta.github.io/cityupdate

I gave deepseek-r1 (q8_0) a math problem and it got there after 10 minutes of non-stop trial and error