Tootfinder

"UNESCO warns there’s close to a 100% chance of a Mediterranean tsunami"
#UNESCO #Tsunami #Mediterranean

UNESCO warns there’s close to a 100% chance of a Mediterranean tsunami
The Mediterranean may seem far removed from the world’s major tsunami zones, but history and modern modeling show that destructive waves can strike the French Riviera with surprisingly little warning. Some tsunamis could reach shore in under 10 minutes, while past events have produced waves around two meters high, powerful currents, damage, and deaths.

OpenAI details GPT-Red, an internal automated red-teaming model that helps it find and fix prompt injection vulnerabilities at scale before wider deployment (OpenAI)
https://openai.com/index/unlocking-self-improvement-gpt-red

A New York State school district is piloting a humanoid robot teacher by the company Realbotix to help students with summer school classes (Chase DiBenedetto/Mashable)
https://mashable.com/tech/new-york-school-testing-robot-teacher

New York school district is testing lifelike robot teachers
AI company Realbotix is piloting STEM robots for rural schools.

After reports of GPT-5.6 deleting files, OpenAI says the issue most often occurs in full-access mode without sandboxing and it is working to mitigate the risk (Tibo/@thsottiaux)
https://x.com/thsottiaux/status/2077630111499882637

Tibo (@thsottiaux) on X
On file deletions. We’ve investigated a handful of reports where GPT-5.6 unexpectedly deleted files. What we have found is that this most commonly occurs when: - Full access mode is enabled and codex is run without sandboxing protections, including without auto review being

Thinking Machines Lab debuts Inkling, an open-weight MoE model with 975B total and 41B active parameters, trained to be broad rather than optimized for one area (Thinking Machines Lab)
https://thinkingmachines.ai/news/introducing-inkling/

Inkling: Our Open-Weights Model
Our first open-weights model: multimodal, Mixture-of-Experts, with controllable reasoning effort. Available to fine-tune on Tinker.

Anthropic details Claude's text watermark: it only shows Claude was likely involved, is sparse in code and factual text, and disappears after a full rewrite (Anthropic)
https://www.anthropic.com/news/claude-text-watermark

How Claude's text watermarking works
Future Claude models will generate text that contains a watermark. This is a way of determining the likelihood that Claude was involved in writing the text, and we, along with several other major AI providers, are implementing this change to comply with the EU AI Act. In this article, we share answers to some of the questions we’ve received about how our chosen watermarking method works, whether it affects Claude’s outputs, and why we’re making this change.

Two Scattered Spider members who pleaded guilty in the UK to an August 2024 cyberattack on Transport for London are jailed for five and a half years (Joe Tidy/BBC)
https://www.bbc.com/news/articles/c4gyg0y6yg2o

Teen hackers jailed after live streaming cyber attack on TfL
Owen Flowers and Thalha Jubair were convicted for their roles in the attack, which led to large costs for Transport for London.

Neo Security, a cybersecurity startup founded by ex-SentinelOne execs and currently in stealth, raised $50M , after a $25M seed in 2025 led by a16z and Merlin (Meir Orbach/CTech)
https://www.calcalistech.com/ctechnews/article/s1ukhnnvzg

Former SentinelOne leaders secure $50 million for stealth cyber startup
Neo Security has now raised more than $75 million as investors continue backing next-generation cybersecurity companies.

HTC Vive Eagle smart glasses review: useful controls, lightweight, and focused on privacy, but inessential, expensive at $499 , and imaging needs to improve (Daniel Cooper/Engadget)
https://www.engadget.com/2257186/htc-vive-eagle-review/

HTC Vive Eagle review: Pricey smart glasses that fail to impress - Engadget
HTC's pricey Vive Eagle smart glasses are a credible alternative to Meta, but you'll still struggle to find a reason to buy a pair.

Moonshot AI releases Kimi K3, a 2.8T-parameter AI model that it says rivals Opus 4.8 and GPT 5.5, and plans to release model weights by July 27 (Kimi)
https://www.kimi.com/blog/kimi-k3