Skip to content
AI Daily · Asia/Shanghai · 2026-11-15 · Sunday· Issue 53· About 8 minutes to read

Science Prediction Universe Benchmarks AI Predictions Against Wet-Lab Results

UniPat AI introduced Science Prediction Universe (SPU), a benchmark that tests whether AI models can predict wet-lab results across chemistry, materials science, biology, and medicine. It includes 100 public tasks drawn from 85 papers and 16 private tasks, covering 4,207 experiments. In an initial evaluation of 12 models across 13 configurations, the highest average score was 0.486, and increasing GPT-6 Astra’s reasoning intensity raised median runtime and API cost without materially changing its score. The dataset and code are open source.
Time machineAcceleration

QbitAIRead the storyOriginal

01

Industry

TechCrunch · AI

Tech Executives Sign Petition Urging Nvidia to Reinstate Jensen Huang

A petition urging Nvidia to reinstate Jensen Huang had attracted more than 115,000 signatures by Saturday afternoon, including 5,965 current employees, 1,108 former employees, and 63,063 shareholders, according to its website. Supporters named in the report include Elon Musk, Amjad Masad, and Tim Sweeney. The petition says Huang was removed without meaningful due process and calls for his return pending an independent investigation.

TechCrunch · AI

TechCrunch Tracks AI Products and Startups That Shut Down or Fell Short

TechCrunch surveys AI products, startups, and initiatives that have shut down, pivoted, or missed expectations, including OpenAI’s ChatGPT redesign rollback, Character.ai’s bankruptcy filing, and Microsoft Recall’s continuing privacy concerns. It cites S&P Global Market Intelligence’s estimate that about 42% of corporate AI initiatives are abandoned, and describes pressures including funding, technical challenges, competition, scaling, and weak user demand.

03

Tips and perspectives

The Verge · AI

Helen Toner on AI Safety, Regulation, and Why AI CEOs Cannot Be Trusted

In a Decoder interview, Georgetown’s Center for Security and Emerging Technology executive director Helen Toner argues that AI safety cannot depend on any one CEO being trustworthy. She discusses how commercial incentives shape AI labs, the limits of alignment and control, and policy options including setting safety thresholds before training or releasing more capable models. Toner says she considers advanced AI plausible enough to prepare for, while cautioning that timelines and outcomes remain uncertain.

Baoyu

How Always-On AI Agents Work: Five Components Behind the Assistant

Baoyu explains an always-on AI assistant through five components: an inbox that wakes it for incoming tasks, timers for proactive actions, memory files, permissions, and an outbox that controls when it speaks. The post says the assistant spends much of its time idle without calling a model or incurring costs, and previews how the setup handles simultaneous tasks, memory limits, approval for important actions, and quiet operation.

QbitAI

How Creator Foyege Made a Harry Potter AI Fan Series

QbitAI profiles foyege, a creator who made an AI Harry Potter fan series that the article says has received 30 million views overseas. It describes the gender-swapped Draco storyline and foyege’s workflow, from selecting a character design to planning storyboards and coordinating Seedance, Wanxiang, and Kling through LibTV’s canvas workspace.

(End of issue)

e/acc DailyAutomatically edited from public sources. Every story links to its original · Daily archive

Tomorrow at 08:00, a briefing like this arrives in your inbox.

The same kind of briefing as the one above: a few stories worth reading. Free. Mail starts after you confirm, and every letter has a one-click unsubscribe. Read a past issue · 2026-11-14