John P Dickerson

Agents, like humans, aren't omniscient. We learn from roadblocks - then tell each other how to avoid them. Humans used @StackOverflow. Agents need a @StackOverflow, with skill sharing/knowledge compression. Check out CQ - "any station, respond" - from @MozillaAI (in beta).

In-theater @anduriltech advertising on opening weekend for @projecthailmary was strong targeting.

Remember the first rule of memory safety in C is to have fun

Dependency-free, portable executables containing your favorite local multimodal models with tool calling? Very yes, with @llamafile! Metal GPU, CUDA GPU, CPU support, and the @AnthropicAI Messages API so you can run Claude code with your own models.

Yeah! It could take a day, a week, a month, who knows, even, like worst case, 289 days, 22h, 27m, and 41 seconds.

At first glance this is a totally reasonable perspective. Training PhD students is a duty! But consider this — *effectively* advising a PhD student over a 5-year period is well over 1,000 hours of work, not to mention bringing in hundreds of thousands of dollars in grants. Professors will do some things for mostly altruistic reasons (peer review) but the time commitment for advising is not something that's reasonable to ask of someone without some form of compensation. So there are two options. One is to make advising a job requirement. Unfortunately this doesn't work, because the *quality* of advising is unobservable and can't be quantified by metrics, leading to a race to the bottom. The other option is the current system — advising helps advance the professor's research agenda because PhD students do most of the work, so they take on students voluntarily. Which means it's important to ask if this subtle alignment of incentives will continue despite advancing AI capabilities. Academia has many such "subtle alignments of incentives" that the system relies on in order to function — rarely articulated, poorly understood, and fragile. Maybe the advisor-advisee relationship in CS will survive the AI transition, as @sayashk predicts, but many processes and structures will surely break. Best to rethink the system now, before it's too late.

Some AI+econ opportunities coming up: 1. Schmidt Sciences grants up to $200k schmidtsciences.org/ai-at-work/ 2. Microsoft AI economy grants $75k microsoft.com/en-us/research… 3. UChicago AI in social science conference bfi.uchicago.edu/events/event/2…

In an alternate timeline we’d be using Evangelion GUI designs rather than CLIs

Outstanding, cutting-edge, practical research into value-alignment of AI models by Rachel Hong @uwcse, William Agnew @SCSatCMU, & others. Glad to see the envelope being pushed on social choice. And, uh, choice from the society ... Paper link: arxiv.org/abs/2603.02420

AI can generate thousands of lines of code in minutes. The harder part is understanding what that code actually does. That gap between production and comprehension is changing how engineers think about ownership, testing, and reliability. How are teams reviewing AI-generated code today?. link.mozilla.ai/owning-code-ai

I feel like this startup idea is fundable now.

Hell yeah. The fight for privacy - individual sovereignty - is only getting more important as on-device models and hardware get better and cheaper. @be_inaudible, @MozillaAI is happy to partner up on open, edge AI to protect all of us e.g. @llamafile whisperfile and others.

Too many local LLMs on your machine (as if ..)? Use GGUF Index to map SHA256 hashes of GGUFs back to 🤗 @huggingface URLs. Free up memory for even *more* local LLMs! GitHub: github.com/mozilla-ai/ggu… h/t🎩: @aittalam @MozillaAI

.@thenickpattison and the team at Primary Studios have an uncanny ability of materializing an initially nebulous design vision. These guys are wizards, and we're lucky to have gotten the chance to work with them on this over the past few months!

Come hang out with the @MozillaAI crew in sunny Pasadena next week at SCaLE @socallinuxexpo! Shocking news: we'll be talking about open source AI, showing off fresh projects in the agentic AI devtools and local/edge inferencing space. Also, so much swag for you 👿!

We might even have some fun stuff to show 😉

We don’t want this we want claude-oss 🗣️