Sign in

William Gunn

@metasynthesis.net
643 followers 465 following 3K posts

Communications and community. More here: synthesis.williamgunn.org/about Talk to me: calendar.app.google/z4KR3xfXTc178es…

PostsRepliesMedia
William Gunn @metasynthesis.net · 01/10/2026
Reading this article was the first time I had heard of the song and I can't believe anyone still thinks there's a single shared cultural experience everyone has. lite.cnn.com/2026/10/01/e...
000
William Gunn @metasynthesis.net · 23/09/2026
Yeah, about those leash-holders...
111
William Gunn @metasynthesis.net · 18/09/2026
000
William Gunn @metasynthesis.net · 17/09/2026
Well done @intelligence.org & @cnn.com ! This is exactly the debate we should be having. lite.cnn.com/2026/09/17/t...
000
William Gunn @metasynthesis.net · 11/09/2026
Can't help wondering what else they'd see if they knew what to look for. lite.cnn.com/2026/09/10/h...
000
William Gunn @metasynthesis.net · 03/09/2026
Man, this season has really jumped the shark. The writing is just over-the-top. lite.cnn.com/2026/09/03/p...
000
William Gunn @metasynthesis.net · 29/08/2026
I'll see your Fiesta and raise you a Copper Pompom (probably not the real name).
000
William Gunn @metasynthesis.net · 26/08/2026
Yet another eminent technology leader sounds a warning on AI. lite.cnn.com/2026/08/26/b...
000
William Gunn @metasynthesis.net · 21/08/2026
Hey @bsky.app, looks like the summary grabbed a redirect message, not the article title & text. Screenshot in case it changes.
000
William Gunn @metasynthesis.net · 14/08/2026
Also, just FYI:
000
William Gunn @metasynthesis.net · 30/07/2026
110
William Gunn @metasynthesis.net · 28/07/2026
If AI is going to take all the jobs, why aren't we seeing it in the data yet? lite.cnn.com/2026/07/28/e...
010
William Gunn @metasynthesis.net · 18/07/2026
Everybody calls this a salty sombrero, right?
000
William Gunn @metasynthesis.net · 13/07/2026
Yet another departure from the safety team at OpenAI. www.businessinsider.com/openai-safet...
000
William Gunn @metasynthesis.net · 11/07/2026
Nice to know my intuition at whatever point in the past was correct.
000
William Gunn @metasynthesis.net · 09/06/2026
I wouldn't call an expanded data retention policy a privacy *protection*, but I get why they do. www.anthropic.com/news/claude-...
010
William Gunn @metasynthesis.net · 07/06/2026
080
William Gunn @metasynthesis.net · 06/06/2026
Have my AI folks been following what's happened to biomedical research funding? diabetesjournals.org/care/article...
000
William Gunn @metasynthesis.net · 27/05/2026
Kudos to Perplexity on this move. OpenAI and Anthropic will need to do something like this too, or they'll lose the researcher demographic, and that will have an outsized impact on adoption throughout industry. www.stm-publishing.com/ebsco-inform...
000
William Gunn @metasynthesis.net · 17/04/2026
We need humans not just "in the loop", but in the design of these AI systems that increasingly affect all our lives. Even with perfect performance, we need transparency to build trust. www.nytimes.com/2026/04/15/m...
010
William Gunn @metasynthesis.net · 13/04/2026
ServiceNow leading the charge to replace workers with AI. www.theregister.com/2026/04/13/s...
000
William Gunn @metasynthesis.net · 10/04/2026
Labor and Capital have co-existed because they need each other. Superintelligent AI threatens to throw off the balance by relieving Capital of its dependence on Labor and neutralizing the power of strikes. However, we can still boycott. www.theguardian.com/commentisfre...
110
William Gunn @metasynthesis.net · 08/04/2026
The whole "develop a dangerously capable AI in private and selectively release it" approach that Anthropic is taking with Mythos is not likely to work sustainably, and once the weights are leaked, the damage is irreversible. lite.cnn.com/2026/04/08/c...
000
William Gunn @metasynthesis.net · 07/04/2026
Why would ScaleAI, a company with Pentagon contracts, be doing this particular task? www.theguardian.com/technology/2...
000
William Gunn @metasynthesis.net · 07/04/2026
Some new info in here, even for people well - briefed on OpenAI history. www.newyorker.com/magazine/202... Yes, people quitting will affect how reckless OpenAI chooses to be.
000
William Gunn @metasynthesis.net · 05/04/2026
The Case For Doing Nothing nosidebar.com/doing-nothing/
000
William Gunn @metasynthesis.net · 28/03/2026
Some good signs for No Kings!
020
William Gunn @metasynthesis.net · 24/03/2026
What's up with the political positioning of the AI companies lately? "It's not (just) about money, it's about power", says David Krueger of Evitable. www.salon.com/2026/03/23/w...
000
William Gunn @metasynthesis.net · 19/03/2026
Apologies for the picture:
101
William Gunn @metasynthesis.net · 17/03/2026
Totally normal technology, just like computers and the Internet, right? www.bbc.com/news/article...
100
William Gunn @metasynthesis.net · 14/03/2026
Funny to think this was only about 12 years ago, now both of those things are trivial. What would be the "research team and 5 years" task today?
010
William Gunn @metasynthesis.net · 10/03/2026
nomadsvagabonds.substack.com/p/the-coalit... On messy coalitions...
000
William Gunn @metasynthesis.net · 07/03/2026
PSA #tmyk
000
William Gunn @metasynthesis.net · 06/03/2026
130
William Gunn @metasynthesis.net · 05/03/2026
That's certainly one take. Here's another.
000
William Gunn @metasynthesis.net · 05/03/2026
The fringes are often the first to speak up, but the centrists won't be far behind once they look at the polling.
100
William Gunn @metasynthesis.net · 05/03/2026
Bernie has his finger on the pulse of society. People want AI that enriches their lives, not something that threatens their job security and human dignity.
000
William Gunn @metasynthesis.net · 05/03/2026
000
William Gunn @metasynthesis.net · 03/03/2026
Technology has improved my life a lot and I'm grateful. Also, this is just gross and not allowed in my house. www.engadget.com/ai/metas-ai-...
000
William Gunn @metasynthesis.net · 26/02/2026
Building an AI-native Research Ecosystem for Experimental Particle Physics: A Community Vision This whitepaper presents a vision for embedding AI end-to-end across experimental particle physics, from detector design through autonomous operations to accelerated analysis. It outlines four grand chal…
Infographic for: Building an AI-native Research Ecosystem for Experimental Particle Physics: A Community Vision
000
William Gunn @metasynthesis.net · 26/02/2026
Three Concrete Challenges and Two Hopes for the Safety of Unsupervised Elicitation This paper identifies three critical challenges for unsupervised elicitation (UE) and easy-to-hard (E2H) generalization methods that steer language models toward truthfulness on tasks beyond human capability: salien…
Infographic for: Three Concrete Challenges and Two Hopes for the Safety of Unsupervised Elicitation
000
William Gunn @metasynthesis.net · 26/02/2026
International AI Safety Report 2026 This is the second International AI Safety Report, synthesizing scientific evidence on general-purpose AI capabilities, emerging risks, and risk management approaches. Led by Prof. Yoshua Bengio with input from over 100 experts across 30+ countries, it assesses …
Infographic for: International AI Safety Report 2026
000
William Gunn @metasynthesis.net · 25/02/2026
I think we can all appreciate this headline.
000
William Gunn @metasynthesis.net · 25/02/2026
In case you didn't know, OpenAI employees can look at everything you send ChatGPT. lite.cnn.com/2026/02/25/p...
000
William Gunn @metasynthesis.net · 14/02/2026
Hot on the heels of OpenAI introducing ads, they have updated their privacy policy and want you to upload your address book (a classic platform growth hack). openai.com/policies/row... Zoe Hitzig may have been on to something...
000
William Gunn @metasynthesis.net · 14/02/2026
Academic publishers are not ready for what's coming. arxiv.org/html/2601.23... Fully-automated math research is here.
000
William Gunn @metasynthesis.net · 10/02/2026
An HBR study of AI at work finds people who adopt it are feeling more stressed. I feel like they have the causality backwards: it's the stress-prone people who are first to adopt it, but if you're not a good driver, buying a Ferrari isn't going to make you better. hbr.org/2026/02/ai-d...
010
William Gunn @metasynthesis.net · 06/02/2026
New polling out of Australia shows AI is a major concern among workers, with 69% wanting better regulation. unionsnsw.org.au/media-releas...
000
William Gunn @metasynthesis.net · 06/02/2026
TakeOverBench.com Mapping the progress to humanity losing control. From @xrobservatory.bsky.social & @pauseai.bsky.social
010
William Gunn @metasynthesis.net · 06/02/2026
Anthropic found that 4.6 surpasses their tests for level 4 (the highest level) autonomy. Instead of taking that seriously, they decided to chuck the eval and just ask people if they should release it instead. 2nd image is an artists depiction of how that went. www-cdn.anthropic.com/0dd865075ad3...
010