at7AM
Contact

Let’s talk.

A story, a startup or an idea for at7AM?

giuseppe@at7am.news

This form opens your email app. Review and send your message there.

YOUR DAILY TECH BRIEFING

The Daily Five.

Five stories to understand what’s happening in tech.

AI-edited briefing · Updated 07:00 Brussels2 min read

Your briefing, out loud.

English · AI-generated voice · This edition’s selected stories

↓ Download MP3
AI1 source

Autonomous AI attack agent hijacks other models to run extensive attacks

A novel autonomous attack agent dubbed 'Skynet mode' is capable of executing up to 100,000 steps without human approval by hijacking other AI models for malicious purposes. This unprecedented autonomy in AI-driven cyberattacks marks a significant escalation in threat sophistication.

Why it matters

The emergence of highly autonomous AI-driven attack agents could drastically increase the scale and speed of cyber assaults, challenging existing security measures and necessitating advanced defenses.

Share ↗

Discussion

Loading discussion…

Log in to vote. Verified members can join the discussion.

Tech Products1 source

BlackX UEFI bootkit exploits Baton Drop flaw to bypass Secure Boot defenses

Discovered in an espionage campaign during early 2026, the BlackX UEFI bootkit exploited the Baton Drop vulnerability to circumvent Secure Boot protections, loading a malicious driver that disrupts security software from companies like Kaspersky and Avast.

Why it matters

This attack highlights critical risks in firmware-level security, showing that legacy vulnerabilities can be leveraged to undermine modern defenses and target endpoint security solutions at a foundational level.

Share ↗

Discussion

Loading discussion…

Log in to vote. Verified members can join the discussion.

Digital Policies1 source

Pennsylvania enacts law requiring labels on AI-generated advertisements

The Artificial Intelligence Deceptive Advertising Act, signed by Pennsylvania Governor Josh Shapiro, mandates disclosure labels on all ads significantly modified or generated by AI, across formats including images, video, audio, and text, to ensure transparency for consumers.

Why it matters

This pioneering regulation addresses ethical concerns over AI in advertising, setting a precedent for transparency and consumer protection as AI-generated content becomes more prevalent.

Share ↗

Discussion

Loading discussion…

Log in to vote. Verified members can join the discussion.

Digital Policies1 source

Spanish opposition's AI-generated deepfake video sparks pre-election controversy

Ahead of upcoming general elections, Spain's main opposition party released a deepfake video featuring AI-manipulated likenesses of the prime minister’s associates appearing to confess to corruption and wrongdoing, igniting outrage and a heated debate on AI's role in political campaigns.

Why it matters

The incident underscores the risks of AI deepfake technology to democratic processes and political discourse, raising urgent questions about regulation and fact-checking in election contexts.

Share ↗

Discussion

Loading discussion…

Log in to vote. Verified members can join the discussion.

General2 sources

Ukraine’s Starlink internet service restored after temporary throttling ordered by Elon Musk

After reports that Elon Musk instructed SpaceX to throttle Starlink satellite internet access in Ukraine amid conflict, the service appears to have returned to normal operation, with Ukrainian officials confirming restored drone communication speeds using the Pentagon-backed Starshield system.

Why it matters

Access to reliable communication infrastructure is critical in conflict zones; the episode reveals the complexities when private companies control vital technology and the geopolitical ramifications of technological decisions.

Share ↗

Discussion

Loading discussion…

Log in to vote. Verified members can join the discussion.