If you had taken the month off and come back today, you would not recognize the AI industry. May 2026 was not a busy news cycle. It was a structural shift. The five biggest things that happened do not fit inside the tech category anymore. They are money stories, labor stories, political stories, distribution stories, and global institution stories that happen to be about AI. Pick any of the last five years of AI coverage and try to find a month where the Pope, a Big 4 firm, the Pentagon, the world's most valuable private company, and 14,000 laid-off workers were all part of the same story. You will not. That month was May 2026. Here is the case, in five receipts, for why it mattered.

The defining number of the month is $900 billion. That is Anthropic's new valuation as of the round that closed this week, up from $380 billion in February. The biggest single-quarter valuation jump for a major private company in the modern venture era. The math behind it is real, not vibes: $11 billion projected Q2 revenue, first operating profit on track, $30 billion-plus round co-led by Sequoia, Dragoneer, Altimeter, and Greenoaks.

Anthropic is not even the only big number. Cerebras went public May 14, raised $5.5 billion in the biggest US tech IPO since Uber, closed up 68 percent at a $66 billion valuation. SpaceX filed for what could be the largest IPO in history with a $1.75 trillion target. OpenAI is preparing its own $852 billion to $1 trillion offering for September. Meta announced $125 to $145 billion in 2026 AI capex, roughly double last year. Anthropic alone is paying SpaceX $1.25 billion a month for GPU compute through 2029.

The era of AI valuations being "investor enthusiasm on vibes" ended in May. These are now public-market-grade numbers, backed by public-market-grade revenue and customer pipelines, about to be tested by actual IPO bookrunners. The grown-up money is here.

On May 19, Andrej Karpathy announced he was joining Anthropic. He co-founded OpenAI. He ran AI at Tesla. He is one of the most recognized researchers in the field. And his stated mission at Anthropic is to use Claude to accelerate the training of the next Claude. AI improving AI as an explicit job description.

Karpathy is the headline but not alone. Anthropic also added Ross Nordeen (founding member of xAI) the same week, plus Chris Rohlf (Meta cybersecurity veteran) to its frontier red team. The pattern across the month: the people who know the most about how this technology actually gets built keep choosing the same side. When the person who literally taught a generation of engineers how neural networks work picks Anthropic over OpenAI, that is not a coincidence. That is a verdict from inside the field.

On May 26, Google Cloud's CEO confirmed publicly what Bloomberg first reported: Apple is paying Google roughly $1 billion a year to license a custom 1.2 trillion parameter Gemini model that will power the next Siri. The full reveal lands at WWDC on June 8.

Read that without flinching. Apple, the company that built its entire brand on doing everything in-house and protecting your data from rivals, decided its own AI was not good enough and rented its assistant's brain from its single biggest competitor. They tested ChatGPT, Claude, and Gemini, and picked Gemini. The model runs on Apple's own Private Cloud Compute infrastructure to preserve the privacy story, but the brain itself is Google.

The strategic read is the part to take away. If Apple, with the largest cash position and engineering bench in the world, concluded that renting was smarter than building on its own timeline, the "build versus buy" question for frontier AI just tipped decisively toward buy. Every other large company is going to follow.

On May 25, Pope Leo XIV released the first encyclical of his papacy. The most authoritative form of teaching document the Catholic Church issues, the kind that shapes the institution's stance for decades. He made it entirely about artificial intelligence and the protection of human dignity. He signed it exactly 135 years to the day after the Church's 1891 document on workers' rights, deliberately framing AI as the Industrial Revolution of our era.

Alongside him on stage was Chris Olah, an Anthropic co-founder, who spoke at the Vatican presentation. To be clear, the Vatican explicitly cautioned that this was not a Church endorsement of Anthropic. But the symbolism was loud. On the same day Anthropic was set to become the most valuable AI lab on earth, its co-founder was standing next to the Pope as a voice for AI ethics, while Anthropic publicly refused the military and surveillance contracts OpenAI has been signing.

When the head of an institution that speaks to 1.4 billion people makes AI his very first priority, AI stops being a tech story. It became a mainstream human one in May.

On May 20, Meta started cutting 8,000 jobs, with another 6,000 open roles canceled, for an effective 14,000-position reduction. About 7,000 surviving workers were reassigned to new AI-focused teams with titles like "AI builder" and "AI pod lead." The cuts happened the same week Meta posted record $56.31 billion in revenue and $26.8 billion in profit.

Intuit announced a 17 percent reduction the same week, also tied to AI. More than 113,000 tech jobs have been eliminated in 2026, many citing AI directly. This is not companies cutting because they are struggling. These are the most profitable firms on earth choosing to spend less on people and more on GPUs, and using AI as both the reason and the tool.

That is the answer to the "will AI take jobs" debate that has been running for two years. Not "AI replaces your task." The actual mechanism: companies reallocating capital from labor to compute, and AI is the lever. Whether that is a smart transition or an overcorrection is still open. But 14,000 Meta workers are not waiting for the debate to settle.

Five receipts. Five categories. One pattern.

The money got real. The talent picked a side. Even Apple tapped out. The Pope made it everyone's story. The labor question got answered. Each of those, on its own, would be a defining moment in a normal year. May 2026 had all five in 30 days.

The pattern underneath them is the same one. AI is no longer being built, sold, debated, or staffed inside the tech industry. It is being built with public-market capital, debated by global institutions, distributed by the largest consumer companies on earth, and staffed by reorganizing the entire labor base of the white-collar economy. The boundary between "AI industry" and "everything else" effectively dissolved in May. The story is now the same story.

That is what "grown up" actually means. Not "more mature in tone." More integrated into the structural fabric of the global economy. The IPOs, the encyclicals, the layoffs, the partnerships, the defections, they all happen in the same month because they are all part of the same shift. Treating them as separate stories is the mistake your competitors are making. They are one story.

The receipts you can already see coming.

WWDC opens June 8. Apple shows the new Gemini-powered Siri in full, the first iPhone era to run on a competitor's brain. SpaceX's IPO roadshow begins June 4, pricing on June 11, trading on Nasdaq under SPCX from June 12, in what could be the largest IPO in history at up to a $75 billion raise. Anthropic confirmed Mythos-class models reach all customers "in the coming weeks," which on the calendar means June. OpenAI is expected to file its own confidential S-1 in June ahead of a September listing.

Four headline events. One month. June is when the moves that defined May actually ship. If May was the month AI grew up, June is the month it goes public, in every sense of the word.

Don’t worry about keeping up, that’s what we’re here for!

May 2026 was the month AI stopped being a tech story and became everything's story. Anthropic closed at $900 billion-plus, the most valuable private AI company on earth, with $11 billion projected Q2 revenue. Andrej Karpathy left OpenAI for Anthropic to use Claude to train the next Claude. Apple is paying Google $1 billion a year to run Siri on Gemini, even Apple could not go alone. The Pope made AI his first encyclical, the strongest signal yet that this is now a mainstream human concern, not a tech one. Meta cut 14,000 jobs the same week it posted record profit, defining what the AI economy actually looks like. Cerebras IPO'd. SpaceX filed. Five receipts, one pattern: the money, the talent, the distribution, the institutions, and the labor base all reorganized around AI in 30 days. June is when the next wave ships. WWDC, SpaceX trading, Mythos for everyone, the OpenAI filing window. The story stops being about whether AI matters. It is now about what kind of world we live in once it is the substrate underneath everything. May was the month that question got answered.

Stay building. 🤖