AI models typically learn by memorizing patterns, then researchers bolt on reasoning as an afterthought. A new method called Reinforcement Pre-Training flips this approach—teaching models to think during basic training instead.
Good Morning from San Francisco, Meta's AI app turned users into accidental exhibitionists. 😬 People shared medical questions, legal
Meta users think they're chatting privately with AI. Instead, they're broadcasting medical questions, legal troubles, and relationship problems to the world through a public feed that many don't realize exists.
A new cybersecurity startup just raised $43 million to tackle a growing threat: AI-powered impersonation attacks. Adaptive Security landed funding from Andreessen Horowitz and OpenAI's startup fund.
The company spotted a critical problem. Attackers now use AI to create fake versions of executives and employees. These AI personas can call, email, or text workers using voice and language that sounds real. The goal? Steal money, data, or plant ransomware.
Brian Long, Adaptive's CEO, builds defenses that fight AI with AI. His platform creates simulated attacks to train employees. When someone gets a fake call from "the controller" asking for a wire transfer, it's actually a test. If they fall for it, they get instant training.
The company's approach works. Since launching in January, they've signed over 100 enterprise customers including the Dallas Mavericks and First State Bank.
For OpenAI, this marks their first investment in cybersecurity. Ian Hathaway from the OpenAI Startup Fund explains why: "AI is reshaping cybersecurity threats faster than most organizations can respond."
Adaptive uses the funding to grow their research team. They're building tools that spot threats in real-time and help security teams act fast. The investment shows the urgent need for better defenses. As AI gets smarter, impersonation attacks grow more convincing. Traditional security training can't keep up with deepfakes that sound and act like real people.
The company's platform goes beyond just attack simulations. It scores risk across teams and creates custom training content. Security teams get a live map of their weaknesses before attackers exploit them.
Why this matters:
Read on, my dear:
Three ex-Cisco executives raised $21M to build AI that handles system outages before engineers get woken up at 2 AM. Their secret: multi-agent systems that cut response time by 90%. But there's a catch that separates them from every other AI monitoring startup.
Perplexity AI, known for providing direct answers instead of traditional search links, is nearing a $14 billion valuation. Yet beneath the soaring numbers lies a pressing question: Can its novel approach truly reshape how we search the internet?
Bugs just got more expensive. Lightrun, an Israeli startup, has secured $70 million to tackle a peculiar side effect of
Benchmark just placed a massive bet on Chinese AI startup Manus, leading a $75 million funding round that values the company at half a billion dollars. Not bad for a startup whose AI agents sometimes crash while ordering takeout.
Get tomorrow's intel today. Join the clever kids' club. Free forever.
