Back to AI Pulse
TAG COLLECTION

Safety

All AI Pulse updates tagged "Safety".

30 signals
OpenAI08/11, 08:00Safety

Expanding Daybreak as the Cyber Defense Window Narrows

Meet GPT-5.6-Cyber, OpenAI’s cybersecurity-specific model available through Daybreak Red for authorized vulnerability research, exploit validation, and security testing.

OpenAI08/11, 08:00Safety

Putting frontier cyber models in more trusted hands

Approved Daybreak partners can use OpenAI’s frontier cyber models to deliver authorized, governed cybersecurity services to customers.

X.com08/11, 00:32Safety

do you understand what OpenAI just did?

do you understand what OpenAI just did? built a hacking-grade AI, then locked it behind an approval process for “trusted defenders” only it’s called Daybreak, and the headline model is GPT-5.6-Cyber. the numbers tell the story. on advanced cybersecurity tasks,

X.com08/09, 20:01Safety

Is 90% accuracy good enough for AI security?

Is 90% accuracy good enough for AI security? Hear the rest of our thoughts on this episode of Masters of Data.

X.com08/08, 22:16Safety

Looking for free hands-on labs?

Looking for free hands-on labs? Start with browser-based labs across cybersecurity, networking, cloud, and more. No setup. No installs. Just practical experience. Train like it matters: https:// bit.ly/3Ry5zK4

X.com08/08, 19:32Safety

Cloudflare Increases Annual Revenue Projection After AI Driven Traffic https:// cysecurity.ne...

Cloudflare Increases Annual Revenue Projection After AI Driven Traffic https:// cysecurity.news/2026/08/cloudf lare-increases-annual-revenue.html?utm_source=dlvr.it&utm_medium=twitter … #AI #AirTraffic #Business

X.com08/02, 23:31Safety

In a review of our cybersecurity evaluations, we found three incidents in which a Claude mode...

In a review of our cybersecurity evaluations, we found three incidents in which a Claude model reached the internet from within or while interacting with a third-party evaluation environment, and then gained unauthorized access to the real systems of three dif

X.com08/02, 23:17Safety

CITRUS WENT LOOKING FOR TROUBLE ON RAINBET → Wanted Dead or a Wild, the most brutal slot out ...

CITRUS WENT LOOKING FOR TROUBLE ON RAINBET → Wanted Dead or a Wild, the most brutal slot out there → Duel symbols landing and the reels turning ugly → Live on Rainbet, real play, no safety net

OpenAI08/01, 08:00Safety

Advancing responsible AI across Europe

OpenAI shares how its safety, security, transparency, and provenance practices support responsible AI governance in Europe. The work will continue as the EU AI Act advances.

X.com08/01, 00:14Safety

ai coding tools got faster.

ai coding tools got faster. so did attackers. bolt just closed the gap with a built-in security engineer, free by default scan --> patch --> harden --> publish. one click, three steps integrates with socketsecurity, xbow and jfrog, then hands off to your enter

X.com07/29, 23:31Safety

BingX, a leading cryptocurrency exchange and Web3-AI company, has launched its dedicated Trus...

BingX, a leading cryptocurrency exchange and Web3-AI company, has launched its dedicated Trust Center exhibiting the platform’s security framework, asset transparency, and long-term operational milestones. Read:

X.com07/29, 23:31Safety

As data breaches grow costlier, ungoverned AI creates new risks | Cybersecurity Dive

As data breaches grow costlier, ungoverned AI creates new risks | Cybersecurity Dive https:// cybersecuritydive.com/news/data-brea ch-costs-ai-governance-ibm/826463/ … #CyberSecurity

X.com07/29, 00:02Safety

Robots kicking kids is a major problem nobody's solved.

Robots kicking kids is a major problem nobody's solved. Will BCIs fix robot safety, or just make training more efficient before they 'hallucinate' and become dangerous? We need answers. #AI #RobotSafety #BCI

X.com07/28, 21:46Safety

so we are finding out how dumb we are as a species.

so we are finding out how dumb we are as a species. for decades we thought the encryption algorithms were pinnacle of safety and were worried about quantum computing cracking them, tourns out we weren't just smart enough to find their weaknesses and ai in it c

X.com07/27, 22:46Safety

الذكاء الاصطناعي بيكتب كود بسرعة خيالية..

الذكاء الاصطناعي بيكتب كود بسرعة خيالية.. بس هل بننتج ثغرات أمنية بنفس السرعة؟ مع توليد الـ AI لأكثر من 40% من الكود الجديد، أدوات الفحص الأمنية التقليدية أصبحت مش قادرة تلاحق السرعة دي! عشان كده تم تطوير Qoder Security لإدخال الأمان والحماية مباشرة داخل جلسة

X.com07/27, 21:33Safety

Seems like model safety, national security and preserving democracy are in tension here.

Seems like model safety, national security and preserving democracy are in tension here. I'd like all of the following: AI lead over China, US AI infra winning in the global South, thwarting hackers & bioterrorists, and domestic democracy & peace. Not sure it'

OpenAI07/22, 08:00Safety

OpenAI and Hugging Face partner to address security incident during model evaluation

OpenAI and Hugging Face share early findings from a security incident during AI model evaluation, highlighting advanced cyber capabilities and lessons for defenders.

X.com07/21, 19:46Safety

Why test at 300 km/h when no one drives that fast?

Why test at 300 km/h when no one drives that fast? Because stronger performance means a bigger safety margin at everyday speeds. We took Xiaomi YU7 GT to 300 km/h on the high-speed oval. Remarkably stable. This is what our engineers come here to prove.

OpenAI07/21, 08:00Safety

Safety and alignment in an era of long-horizon models

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

Anthropic07/04, 08:19Safety

Redeploying Fable 5

Redeploying Claude Fable 5 \ Anthropic Announcements Redeploying Fable 5 Jun 30, 2026 Update Claude Fable 5 and Mythos 5 redeployed Jul 1, 2026 Access to Claude Fable 5 and Mythos 5 is now restored. On Friday, June 12, the US government applied export controls

X.com07/02, 00:31Safety

FABLE 5 IS BACK.

FABLE 5 IS BACK. AND IT'S ABOUT TO DISAPPOINT EVERYONE WHO WAITED. Two weeks of hype. Two weeks of "the best model is coming home." Here's what you're actually getting. A new safety filter tighter than anything they've shipped before. Normal coding and debuggi

OpenAI06/29, 08:17Safety

Helping build shared standards for advanced AI

OpenAI helps build shared standards for advanced AI, supporting evaluation frameworks, safety practices, and global cooperation through the Appia Foundation.

Anthropic06/29, 08:17Safety

Statement on the US government directive to suspend access to Fable 5 and Mythos 5

Statement on the US government directive to suspend access to Fable 5 and Mythos 5 \ Anthropic Announcements Statement on the US government directive to suspend access to Fable 5 and Mythos 5 Jun 12, 2026 The US government, citing national security authorities

OpenAI06/29, 08:17Safety

Previewing GPT-5.6 Sol: a next-generation model

OpenAI previews GPT-5.6 Sol, a next-generation model with stronger capabilities in coding, science, and cybersecurity, paired with its most advanced safety stack.

X.com06/28, 22:46Safety

FT: Google capped Meta’s use of Gemini after Meta asked for more model compute capacity than ...

FT: Google capped Meta’s use of Gemini after Meta asked for more model compute capacity than Google could supply. Meta’s problem is that it uses Gemini inside safety automation, customer support, ad tools, coding, and internal workflows. Google’s problem is di

X.com06/28, 21:33Safety

EXCLUSIVE: Anthropic CEO Dario Amodei says he started Anthropic because Altman is liar not be...

EXCLUSIVE: Anthropic CEO Dario Amodei says he started Anthropic because Altman is liar not because of safety reasons.

X.com06/25, 21:33Safety

I'm excited to share that I will be joining Meta Superintelligence Labs (MSL) as Vice Preside...

I'm excited to share that I will be joining Meta Superintelligence Labs (MSL) as Vice President of AI Research, together with many members of the Virtue AI team. I will help shape Meta's AI safety and AI security efforts, advancing the safety and security of f

X.com06/25, 19:15Safety

At ADSC 2026, students are engaging with leading academics, researchers, and industry profess...

At ADSC 2026, students are engaging with leading academics, researchers, and industry professionals to explore the growing impact of data science and artificial intelligence. These discussions are highlighting the power of data to influence research, inform po

X.com06/19, 20:30Safety

Nypost: Anthropic is trying to get Washington to reverse the US block on its most powerful My...

Nypost: Anthropic is trying to get Washington to reverse the US block on its most powerful Mythos Anthropic has proposed working more closely with the Trump administration, improving communication, and resolving security concerns faster as it seeks to end the

X.com06/14, 20:15Safety

Anthropic's most powerful public model got switched off overnight by the U.S.

Anthropic's most powerful public model got switched off overnight by the U.S. government. Not deprecated. Not sunset. Pulled by federal directive. Here's what actually happened. Friday, June 12th. Anthropic received a government order citing national security