53rd Edition Download

Get answers in real-time with the AI Interview Buddy on iOS.

This Week in AI:

From big-stage keynotes to billion-dollar startups, this week’s AI moves are loud, strategic, and hard to ignore.

Google’s I/O 2025 is going to be a full-court press for AI everything: Gemini, Android, and new dev tools will take center stage. OpenAI launched HealthBench, a new benchmark to evaluate AI in healthcare reasoning. And Perplexity just hit a $1.4 billion valuation.

Here’s what it all means and why it matters to you.

Let's dive in.

In This Issue:

• Google I/O 2025 → This year is adding up to be all about AI (link)

• OpenAI Launches HealthBench → A new benchmark to test AI in healthcare settings. (link)

• Perplexity Hits $1.4B Valuation → AI search is hot again and investors are all in. (link)

Google I/O 2025

TL;DR:

Google I/O kicks off next week, and it’s expected to be one of the most AI-packed keynotes yet. Gemini is rumored to get major upgrades, and leaks point to new Android integrations, real-time video tools, and maybe even an AI-powered hardware assistant.

Our Take:

For AI-curious readers and builders, this event is a temperature check on Google’s direction: are they still playing catch-up with OpenAI and Anthropic, or are they ready to lead? Keep an eye on developer tools, Gemini-powered search features, and any moves that tighten Gemini into Android’s DNA. If you work in the AI ecosystem at all, as a dev, founder, or just trying to stay ahead of the curve, what drops this week could directly impact what you build (or compete with) in the next 6 months.

OpenAI Launches HealthBench

TL;DR:

HealthBench is OpenAI’s new benchmark for evaluating AI models in real-world healthcare settings. Built with input from 262 physicians across 60 countries, it features over 5,000 realistic health conversations, each with custom rubrics based on expert criteria. The goal? Measure AI in ways that reflect how doctors actually think, not just test scores or chatbot fluff.

Our Take:

This is OpenAI going deep, not wide. HealthBench isn’t just a dataset, it’s a move to redefine how we evaluate AI in critical domains like medicine. And it shows OpenAI’s serious intent to move LLMs from helpful to clinically trusted. For readers working in healthtech, research, or even startups building agentic systems, this benchmark is something to study. It sets a new bar for safety, transparency, and rigor. And here’s the kicker: their latest models now outperform physicians in many of these tasks, but not all. There’s still headroom, and that’s the point.

Perplexity Hits $1.4B Valuation

TL;DR:

Perplexity, the rising star in AI-powered search, just closed a funding round that pushed its valuation to $1.4 billion. Known for delivering clean, citation-backed answers, it’s been steadily winning over users (and investors) looking for a more transparent alternative to Google’s increasingly noisy results.

Our Take:

This isn’t hype. It’s the kind of product-market fit AI founders dream of. Perplexity is riding two massive waves: (1) distrust in traditional search UX, and (2) the demand for speed + trust in AI answers. If you’re building anything that touches research, search, or knowledge delivery, Perplexity’s rise is your signal. Search is still being rewritten, and people are hungry for a smarter, cleaner interface. You might want to beat them… or build on top of them.