There are weeks where the news cycle merely hints at civilizational inflection points, and then there are weeks where a Nobel laureate proposes an intelligence test named after Einstein, a nonprofit becomes worth more than most countries’ GDP, and the sun and wind quietly dethrone the fuel that powered the last century. Peter Diamandis’s EP #260 of Moonshots arrived on June 1st like a sommelier presenting a flight of existential milestones — each one paired with a vintage sense of “well, that escalated quickly.”

We at The Slap of Wisdom have done what any self-respecting luxury commentary publication would do: we tasted every glass, spat out the hype, and kept the facts. You’re welcome.

🤚 The Open-Palm Inventory of Inflection Points

Let us begin with the headline that launched a thousand benchmark arguments. Anthropic’s Claude Opus 4.8 now sits atop the Artificial Analysis Intelligence Index at 61.4, a full 1.2 points ahead of GPT-5.5. On SWE-Bench Pro — the coding benchmark that actually matters — Opus 4.8 scored 69.2 versus GPT-5.5’s 58.6. That’s not a margin. That’s a different zip code.

But the episode’s real spine-tingling moment wasn’t about which model writes better Python. It was about Demis Hassabis, CEO of Google DeepMind, Nobel Prize winner, and the closest thing the AI industry has to a philosopher-king, aligning his AGI prediction with Ray Kurzweil’s 2029 timeline. Hassabis described humanity as standing in the “foothills of the singularity” — which is either deeply poetic or deeply terrifying, depending on whether you own Nvidia stock.

Meanwhile, in the quiet revolution no one talks about at dinner parties: wind and solar generated more electricity than natural gas globally for the first time ever in April 2026. According to Ember, the two sources produced 531 TWh (22% of global electricity) versus gas at 477 TWh (20%). Five years ago, gas was generating nearly double what wind and solar could manage. The exponential curves Peter keeps preaching about? They showed up in the energy grid while you were arguing about chatbot benchmarks.

👐 The Two-Handed Philanthropy Paradox

Now let us discuss the most quietly extraordinary development of the week: the OpenAI Foundation.

Through OpenAI’s corporate restructuring, the nonprofit Foundation now holds a 26% equity stake in OpenAI’s for-profit entity — a position valued at roughly $130 billion at the time of restructuring, and potentially worth $220 billion or more at current valuations. This makes the OpenAI Foundation, by a comfortable margin, the largest charitable foundation on planet Earth. It surpasses the Gates Foundation, the Ford Foundation, and every sovereign wealth fund that dabbles in altruism.

And what does the world’s richest charity plan to do with its windfall? The Foundation has committed $250 million to workers displaced by the very technology it helped create. It is funding research into public wealth funds and AI dividends — concepts that sound like science fiction until you remember that the entity writing the checks literally invented the technology that’s making the jobs disappear.

There is something profoundly 2026 about a company building the automation machine and writing the severance checks. The phrase “creating the problem and funding the solution” has never had a more expensive implementation.

🌿 The Gentle Awakening — Or, What Would Einstein Think?

Back to Hassabis and his proposed “Einstein test” for AGI, because this deserves its own section and possibly its own wing in a museum of intellectual audacity.

The test works like this: take a frontier AI model, restrict its training data to everything published before 1901, and then ask it to independently derive special relativity — the insight Einstein produced in 1905 that fundamentally rewrote physics. If the model can do it, you’ve got something that can genuinely reason rather than merely recombine.

Today’s systems cannot pass this test. Not Opus 4.8, not GPT-5.5, not Gemini 3.5. None of them. And Hassabis is refreshingly honest about this, which is more than you can say for most people who use the word “AGI” in a sentence.

What makes the Einstein test elegant is that it doesn’t measure how well a model has memorized the internet. It measures whether a machine can do what the greatest human minds did: look at the evidence available and make a cognitive leap that nobody else saw coming. It is, in Hassabis’s framing, the difference between an incredibly sophisticated library and an actual thinker.

He believes we’ll get there by 2029. Kurzweil agrees. The rest of us are standing in the foothills, checking our step count.

👑 The Crown Verdict — Quantum Chips and the Week That Broke the Scale

As if the episode needed one more civilizational data point, Diamandis also covered IBM and the Department of Commerce announcing Anderon — a purpose-built quantum chip foundry in Albany, New York, backed by $2 billion ($1 billion CHIPS Act, $1 billion IBM cash). The facility will manufacture quantum devices on 300mm wafers, producing chips 30 times faster than current methods, operating 24/7 with full automation.

Let us pause and appreciate the ambition: the United States is building a factory to manufacture the physical substrate of post-classical computation. It sounds like the plot summary of a civilization-building video game, except the budget is real and the factory will actually run around the clock.

And then there’s Amazon, whose AI shopping assistant now converts customers at 3.5 times the rate of traditional keyword search. The company plans to license this capability to other retailers via an AWS-style platform. Because of course Amazon would find a way to monetize the act of making you spend money faster. The marketplace is eating itself, and the ouroboros has an API.

This is the week Peter Diamandis summarized: the smartest model yet, the richest charity yet, the first time the sun outpowered gas, a quantum factory, and an intelligence test designed by a Nobel laureate named after a patent clerk. If you’re keeping score at home, the singularity isn’t arriving — it’s furnishing the apartment.

Inspired by Opus 4.8 Beats GPT 5.5, the $220B OpenAI Foundation, and Hassabis’s 2029 AGI Prediction | EP #260 by Peter Diamandis.

Your inflection point is showing. Extrapolate wisely.