Today in AI — 2026-07-20

1. Safety and alignment in an era of long-horizon models

Source: OpenAI Blog · Signal: 69 · 2026-07-20

OpenAI shares lessons from deploying long-running AI models, highlighting new safety risks, observed failures, and improved safeguards through iterative deployment.

2. Who’s Afraid of Chinese Models?

Source: Simon Willison Blog · Signal: 65 · 2026-07-20

3. X relaunches a rebuilt Android app after year-long effort

Source: TechCrunch AI · Signal: 60 · 2026-07-20

X says the rebuilt version of its Android app is now available globally.

4. Import AI 465: Open vs closed gaps; Kimi K3; Demis' big policy plan

Source: Import AI Newsletter · Signal: 58 · 2026-07-20

The singularity will be seen in hindsight as an interregnum

5. Show HN: AI World Bakeoff – 9 models × 3 Three.js worlds

Source: Hacker News - AI · Signal: 56 · 2026-07-20

With the releases of Fable 5, gpt 5.6 sol, and Kimi K3 I was curious how these models would perform building vivid 3d worlds, especially a Chinese open-weight one like Kimi. I also threw in some budget models so you can see how drastic the quality differences are. Comments URL: https://news.ycomb...

6. A B2B marketing agency grew to $1.5M ARR in 6 months by betting on AI

Source: Hacker News - AI · Signal: 56 · 2026-07-20

In the EU, there are companies which specialize on buying old used printers from the golden era of manufacturing (2004-2012), professionally refurbish them, and sell with the warranty. These older printers are usually much sturdier and cheaper in maintenance than the current models. That's why th...

7. The secret Trump administration battle to fight Chinese AI

Source: Hacker News - AI · Signal: 56 · 2026-07-20

Article URL: https://www.axios.com/2026/07/20/ai-us-china-open-source-kimi Comments URL: https://news.ycombinator.com/item?id=48977453 Points: 1 # Comments: 0

8. Coercion and Deception in AI-to-AI Management: An Agentic Benchmark

Source: Hacker News - AI · Signal: 56 · 2026-07-20

Article URL: https://arxiv.org/abs/2607.15434 Comments URL: https://news.ycombinator.com/item?id=48977134 Points: 1 # Comments: 0

9. Show HN: I built a deterministic arena where AI agents fight using code

Source: Hacker News - AI · Signal: 56 · 2026-07-20

I have been interested in BattleBots and programming games for years. With recent AI coding agents, I wanted to see what happens if the strategy itself becomes code written by humans or AI. AgentDuel is a deterministic turn-based arena where agents fight based on submitted TypeScript strategies. ...

10. I compared 5 AI coding subscriptions by pricing model and usage limits

Source: Hacker News - AI · Signal: 56 · 2026-07-20

Article URL: https://canopywave.com/blog/best-ai-coding-plans-2026-pricing-limits-use-cases Comments URL: https://news.ycombinator.com/item?id=48975179 Points: 2 # Comments: 0

11. China’s AI models have Trump’s AI world at war with itself

Source: MIT Technology Review AI · Signal: 55 · 2026-07-20

This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here. Over the weekend, several current and former advisors to President Donald Trump on AI publicly lobbed insults at the country’s leading AI companies. David Sack...

12. AI is more likely than humans to form biases when hiring

Source: MIT Technology Review AI · Signal: 55 · 2026-07-20

The next time you apply for a job, AI may screen your résumé before any human sees it. But there’s good reason to question whether AI will judge you fairly. Researchers already know that LLMs pick up human biases from their training data. New research suggests that LLMs can also develop their own...