Who's Who of AI

Joshua Grochow

70 trust researcher @joshuagrochow.bsky.social · 2,468 followers
Research: TheoryCompSci, pure math, complex systems Other: climate; covid; equity, inclusion, & accessibility Assoc. Prof. @ CU Boulder Comp. Sci. & Math Views my own @[email protected] (& prev twitter) https://home.cs.colorado.edu/~jgrochow
What they're sharing

Articles & links

↻ Joshua Grochow reposted
Suresh Venkatasubramanian @geomblog.bsky.social

Almost 30% of the submissions to the Neurips position paper track were deemed to be ai generated or heavily ai-implicated. blog.neurips.cc/2026/06/02/a...

AI-Generated Papers in the NeurIPS 2026 Position Paper Track – NeurIPS Blog blog.neurips.cc
AI Weekly's analysis →
  • NeurIPS desk-rejected 178 papers (18.4%) via Pangram; 123 more must prove human authorship by June 15, 2026.
  • Default Pangram window sizes flagged 42.7% of submissions; refined 100-word windows reduced that rate to 12.7%.
  • Comparison venues FAccT 2025 and NeurIPS E&D 2026 showed 0% and 2.1% maximum AI scores respectively.
Read full analysis →
 View on Bluesky →
↻ Joshua Grochow reposted
404 Media @404media.co

NEW: Companies are deliberately making their AI tools speak like cavemen in an attempt to stop burning through AI tokens and curb their massive expenditure on AI, 404 Media has found. “Caveman save you token, save you money.”

Companies Are Making Claude and Codex Talk Like Cavemen to Stop AI’s Soaring Costs 404media.co
AI Weekly's analysis →
  • A plugin called caveman, written by Julius Brussee in early April, strips verbose model output and cut tokens by roughly 65 to 75 percent in his tests.
  • Shayne Sweeney, OpenAI's director of engineering, contributed code to caveman to support Codex, and developers at Nvidia and GitHub are reportedly using it.
  • GitHub shifted to per-token billing in April, Uber blew through its entire AI budget in four months, and Legrand's internal memo points staff at caveman.
Read full analysis →
 View on Bluesky →
↻ Joshua Grochow reposted
@caulfieldtim.bsky.social

⚠️ Another problem with AI: makes us worse at spotting bunk! The consequences of relying on AI for accurate news news.mit.edu/2026/consequ... "...participants who relied on AI systems to verify facts actually got worse at detecting misinformation on their own when their chatbo…

The consequences of relying on AI for accurate news news.mit.edu View on Bluesky →
↻ Joshua Grochow reposted
@schnizzl.bsky.social

Can LLMs Emulate Human Belief Dynamics? 🧪 The answer is a clear no. [The findings] raise a sharp warning against deploying LLMs as human proxies in social simulations. arxiv.org/pdf/2605.18781 #SyntheticData #EsWirdImmerAbsurder

arxiv.org View on Bluesky →
Their own posts

Recent commentary

Large language model the size of a small language model

View on Bluesky · ♥ 9 ↻ 0 ↩ 0 · 57d ago

Is it possible to add language to a copyright notice so that if someone knowingly feeds your work into an LLM as part of a prompt (not training) that it would infringe the copyright? Assume for the sake of discussion that using your work to train an LLM w/o your consent is a violation of copyright.

View on Bluesky · ♥ 6 ↻ 0 ↩ 1 · 30d ago

Thanks genAI companies. I really appreciate that every website now has a "verify you are human" captcha. New badness: b/c of this, many tabs no longer stay cached, they revert to "confirm form (re)submission" (the captcha form). LLMs have definitely improved the internet in so many ways.

View on Bluesky · ♥ 6 ↻ 0 ↩ 0 · 15d ago

Programming is like what magic is supposed to be: you write just the right archaic symbols and *something very specific happens* (but be careful what you wish for etc.). LLMs are a bit more like the everyday experience with "magic": sleight of hand (can be very impressive or very disappointing).

View on Bluesky · ♥ 5 ↻ 0 ↩ 0 · 17d ago
Their network

In Joshua Grochow's orbit

Center = Joshua Grochow. Left = members they follow (green edges). Right = members who follow them (blue edges). Top = mutual follows (orange edges, slightly larger). Drag any node to reposition; click to open that profile.