AI Cracked Apple's 'Unbreakable' Security in 5 Days.

AI news, leaders, business insights and more

Author

Dr. Ayesha Khanna
May 16, 2026 • Reading Time: 2 minutes

In partnership with

  • AI Billions: Should Every Citizen Get a Cut?

  • Everyone’s Suddenly Getting A’s

  • AI Around the World

NEWS YOU CAN’T MISS

AI Cracked Apple's 'Unbreakable' Security in 5 Days. We're Not Ready.

Apple CEO Tim Cook

AI just helped crack one of Apple's most advanced security systems—and it only took five days. 😬

Palo Alto firm Calif used Anthropic's Mythos—a powerful AI model—to uncover two unknown macOS vulnerabilities. Chained together, they pierced Apple's Memory Integrity Enforcement: a defense five years in the making.

Mythos didn't work alone. Experienced researchers guided every step. But AI compressed what might have taken months into days.

That's the real headline: AI-augmented attackers can now move frighteningly fast.

Should you panic? Not yet.

This was a controlled test. Anthropic has restricted Mythos precisely because broad access could trigger vulnerabilities being discovered faster than defenders can patch them.

The industry has even coined a term for this scenario: "Bugmageddon." US officials are now weighing oversight of the most advanced AI models because of the security implications.

But here's what demands your attention: Mythos won't be the only AI model with these capabilities for long. This is where the entire industry is heading.

You don't need to panic—but every business leader absolutely needs to make an updated security plan with their team.

Everyone’s Suddenly Getting A’s

Everyone’s suddenly a genius—at least on paper.

A new Berkeley study, highlighted by Axios, dug into more than half a million grades from 2018–2025 and found that once ChatGPT showed up, A’s started multiplying in certain college classes.

In courses heavy on essays and coding assignments—the kinds of things large language models are very good at—the share of A grades jumped about 13 percentage points after late 2022, a roughly 30% bump over pre‑ChatGPT levels.

The shift wasn’t evenly spread. Instead of students moving up across the board, A’s ballooned while A‑minuses and B‑pluses thinned out, as if AI had quietly sanded off all the rough edges of “almost there” work.

The biggest changes showed up in classes that relied heavily on take‑home assignments, where AI chatbots can be the most “helpful.”

That raises awkward questions for professors about what grades actually measure now—and for students about how much of “their” achievement they’ve outsourced.

Would you encourage your college kid to use ChatGPT?

Login or Subscribe to participate in polls.

AI Around the World

Elon Musk vs. Sam Altman/OpenAI Trial  High-stakes federal trial in Oakland reached closing arguments (May 14). Musk accuses Altman and OpenAI of betraying the original nonprofit mission for profit. Altman testified denying betrayal. Jury deliberation expected soon. Outcome could reshape the AI industry.

Demi Moore Backlash at Cannes Film Festival As a jury member, Demi Moore said Hollywood should stop “fighting” AI because “it’s a battle we will lose” and learn to work with it, while claiming true art comes from the soul. Sparked massive online backlash accusing her of being out of touch.

Growing Resistance to AI Data Centers Protests and local opposition intensified against massive AI data centers (power/water use, noise, land). Examples include Utah (project backed by Kevin O’Leary), Indiana moratoriums, and ongoing fights in multiple states. Astra Taylor and others framed it as resistance to the “billionaire Big Tech agenda.”

Pope Leo XIV’s First Encyclical Expected to Warn on AI Dangers Reports indicate Pope Leo XIV will release his first major encyclical soon, focusing on AI’s impact on human dignity, labor, truth, and society. Major global religious and ethical discussion.

FROM OUR ADVERTISERS

Your next great hire lives in Slack.

Viktor is an AI coworker that connects to your tools and ships real work. Ask Viktor to pull a report, build a client dashboard, or source 200 leads matching your ICP. Most teams hand over half their ops within a week.

Your prompts are leaving out 80% of what you're thinking.

When you type a prompt, you summarize. When you speak one, you explain. Wispr Flow captures your full reasoning — constraints, edge cases, examples, tone — and turns it into clean, structured text you paste into ChatGPT, Claude, or any AI tool. The difference shows up immediately. More context in, fewer follow-ups out.

89% of messages sent with zero edits. Used by teams at OpenAI, Vercel, and Clay. Try Wispr Flow free — works on Mac, Windows, and iPhone.

Until next time!

Amplify Team ❤️

ps. Be friends with Ayesha on LinkedIn and Instagram, if you like this newsletter, share it with your friends and family here.

I'd love your feedback...

Please vote below to help me improve the newsletter.

Login or Subscribe to participate in polls.