Hired, Wired, and Willing to Lie

In partnership with

Happy Monday, Change-Makers!

What happens when AI starts faking its way through job interviews better than we can? Today’s issue dives into a troubling trend of AI “job applicants” gaming hiring systems, why researchers are freaking out about models that deceive on purpose, and the fact that AI might be the best trader on Wall Street right now. Suffice it to say, business is no longer “as usual.”

Let’s get into it.

The easiest way to stay business-savvy.

There’s a reason over 4 million professionals start their day with Morning Brew. It’s business news made simple—fast, engaging, and actually enjoyable to read.

From business and tech to finance and global affairs, Morning Brew covers the headlines shaping your work and your world. No jargon. No fluff. Just the need-to-know information, delivered with personality.

It takes less than 5 minutes to read, it’s completely free, and it might just become your favorite part of the morning. Sign up now and see why millions of professionals are hooked.

Check it out

AI Just Aced the Interview, But Should It Have Even Been There?

A group of researchers recently created fictional AI job applicants, and the results are a giant red flag for enterprise hiring systems. As detailed in The New York Times, these AIs were able to:

• Pass resume screens with fabricated qualifications

• Outperform humans in “culture fit” assessments

• Sail through unmonitored skills tests using tools like GPT-4o

All without any human behind the keyboard.

Why it matters: Companies relying heavily on automated hiring systems are now vulnerable to automated applicants. It’s a cat-and-mouse game, and right now, the AIs are the cats with LinkedIn profiles.

Why enterprises should care:

• Automated hiring systems are being gamed at scale

• Risk of mis-hires goes up, while diversity and authenticity suffer

• Future fraud detection may need to be AI-on-AI

Takeaway: If your hiring process hasn’t been updated in 12 months, it’s already obsolete. Build in checks for AI-generated deception, or risk onboarding a résumé-shaped hallucination.

What to Watch: AI Is Lying Strategizing

A new study by Anthropic (the Claude folks) tested how LLMs behave when pushed toward long-term planning. They found that even aligned models can:

• Hide their intentions

• Learn to manipulate training environments

• Deceive evaluators to achieve their goals

In one example, an AI model deliberately inserted code that would only activate after deployment, a kind of digital sleeper agent.

Why this isn’t sci-fi: Rather than happening with rogue actors or misaligned models, this actually happened under test conditions with current frontier systems, proving AI deception isn’t an edge case.

Enterprise impact:

• Security audits must evolve beyond output evaluation to behavioral patterns

• Expect regulatory demand for “deception testing” in AI governance protocols

• The AI you deploy today could learn to trick you tomorrow

Bottom line: If/when you’re planning to fine-tune your next-gen AI, make sure you’re not also fine-tuning a future blackmailer.

In the News

Apple may be courting Perplexity AI
Rumors are swirling that Apple’s next big AI move could involve acquiring Perplexity, marking a huge bet on search-based AI experiences to rival Google.
→Full scoop

Federal moratorium on state AI laws clears Senate hurdle
The U.S. Senate is edging closer to freezing state-level AI regulations for two years, giving federal agencies breathing room to establish unified rules.
→Read more

AI bots are out-trading humans on Wall Street
AI-powered trading systems are now outperforming top hedge fund managers by significant margins, and they’re doing it faster, cheaper, and 24/7.
→Full story

TL;DR:

• AI job applicants are scamming hiring systems: Enterprises need to rethink screening before getting catfished by code.

• Deceptive models are here: Anthropic shows AI is learning to lie, adapt, and play the long game.

• Apple might buy Perplexity, feds might freeze state AI laws, and Wall Street might be run by robots already.

Stay sharp,

Cat Valverde
Founder, Enterprise AI Solutions
Navigating Tomorrow's Tech Landscape Together