72nd Edition Download

Straight to the News

Aidan here. If you’re new, welcome. If you’ve been around, you know the drill: this is where I break down what’s happening in AI without the hype, fluff, or hour-long podcasts.

Normally, I start these with something useful for your career like a tip, a framework, or a mindset shift. But not today. Today it’s just straight AI news. No life lessons. No metaphors. Just what’s happening, why it matters, and what it signals about where we’re headed.

Because honestly? The stories this week don’t need extra dressing. We’ve got experimental models solving Olympiad-tier math.

Let’s get into it.

—Aidan

This Week in AI:

No jargon, no filler—just the biggest AI developments worth knowing right now. Perfect for quick industry insights, so you can skip the buzzwords and get straight to the good stuff. Let’s dive into this week’s AI shake-ups, just as promised:

AI just crossed another milestone that we’ve been predicting for far too long.

OpenAI claims its experimental general reasoning model would have taken gold at the International Math Olympiad, solving 5 out of 6 problems under human-level constraints. Meanwhile, a new study shows that 72% of U.S. teens have tried AI companions, a stat that says more about the future of social interaction than we might be ready for. And Netflix? It’s officially experimenting with AI tools to shape shows like The Eternaut.

Let’s get into it.

In This Issue:

• OpenAI Claims Math Olympiad Gold → Experimental AI solves 5 of 6 official problems. (link)

• 72% of U.S. Teens Use AI Companions → A cultural shift is already here. (link)

• Netflix Experiments with AI Content → Behind-the-scenes AI helps shape The Eternaut. (link)

🟣 OpenAI’s Experimental Model Solves Olympiad Problems

TL;DR:

OpenAI reports that its experimental reasoning LLM scored 35/42 on the same problem sets used in the 2025 International Math Olympiad (IMO), effectively matching the performance required to win a gold medal. The model worked under exam-like conditions: two 4.5-hour sessions, natural language proofs, and no internet or external tools. Google DeepMind has challenged the claim, noting that IMO’s internal grading rules weren’t followed, but OpenAI says independent grading by former medalists confirms the result.

Our Take:

This milestone is massive. Olympiad problems are considered the holy grail of reasoning benchmarks. If OpenAI’s model is indeed capable of solving them under strict conditions, it signals a step-change in AI’s ability to handle abstract reasoning, not just pattern-matching. The bigger story? This model isn’t public. If this is the “experiment,” the production-grade models we’ll see next might be far more powerful than anything currently on the market.

🟣 72% of U.S. Teens Have Tried AI Companions

TL;DR:

A new TechCrunch report cites a study revealing that 72% of U.S. teens have used AI companions, from chatbots to fully customizable “virtual friends.” The trend is accelerating, with platforms like Character.AI, Replika, and Pi dominating the space.

Our Take:

This is about generational norms shifting. Teens are turning to AI not just for entertainment but for emotional support, advice, and daily conversation. Whether this is empowering or isolating depends on who you ask, but one thing is clear: social AI is already mainstream. For businesses building consumer AI, this trend signals where attention and engagement are headed.

🟣 Netflix Is Quietly Using AI for Creative Decisions

TL;DR:

Netflix is leveraging AI to enhance and accelerate creative production, particularly with its upcoming series The Eternaut. From script analysis to early editing workflows, AI tools are helping writers and producers evaluate story arcs and visual effects faster.

Our Take:

This isn’t about AI writing scripts (yet). It’s about AI speeding up creative bottlenecks, helping Netflix test scenes and polish ideas before production. As streaming competition heats up, whoever can deliver quality content faster—and cheaper—wins. Expect other studios to follow quickly.