For years, Anthropic has sold itself as the responsible one. It was the lab founded by people who left OpenAI over safety worries, incorporated as a public benefit corporation, with a special trust meant to keep the mission ahead of the money. Now it has filed to list on the Nasdaq under the ticker ANT. The company that said it was building a dangerous technology carefully is about to answer to quarterly earnings calls.
Start with the numbers, because they are bizarre. Anthropic reports $16.3 billion in revenue for the first half of 2026, up from $1.2 billion a year earlier. It also reports a net loss of $80.7 billion, up from nearly $13 billion. That is about five dollars lost for every dollar earned. Chips and infrastructure cost $8.9 billion, which explains only a fraction of the hole. The filing itself says revenue growth will slow, and it flags reliance on debt and on training new models with old ones as risks. So the company is telling investors the growth curve bends while the losses are measured in the tens of billions. Someone should demand a line-by-line account of what fills that gap before ordinary people are invited in.
Because that is what an IPO is. Early backers, the venture funds and the corporate giants who fed this machine, get to turn paper valuations into cash. The risk then passes to whoever buys at the open: retail traders, pension funds, and eventually anyone whose retirement account tracks an index. Private capital takes the upside while the public absorbs the exposure. We have watched this movie, and it does not always end well for the people buying in last.
Then there is the question of what the revenue is. A thirteenfold jump in six months did not come from nowhere. Much of it is businesses paying to have work done by software that once went to people. Executives know this, and it is the pitch. The Left should not be Luddite about it. Technology that cures disease or cuts drudgery is worth wanting. But a company that grew this fast on the accumulated writing, code and labor of millions of people, and that has already paid for the privilege of settling with authors over pirated books, owes more than a mission statement when it tells the public what comes next. Who gets the gains from automating the economy? So far the answer is the cap table.
This is also where my own side should be embarrassed. Democrats have spent years courting AI money, praising American "leadership," and promising frameworks and summits while no binding rules arrived. Meanwhile data centers are driving up demand on power grids that ratepayers fund. We let the industry write its own safety standards and called it partnership. An SEC filing is, perversely, the most serious disclosure regime AI has faced.
The central test is governance. Public benefit status and a long-term trust sound reassuring until a public company misses numbers and activist investors start asking why safety research is eating margin. Fiduciary pressure is not a rumor. It is the logic of the market Anthropic is joining. If the company's safety commitments are real, they should be written into binding, enforceable terms that survive a bad quarter, with disclosure about what the trust can actually veto and who sits on it. If they cannot survive that scrutiny, we have learned something important about what "responsible AI" meant.
Regulators should read this filing the way they would read a bank's. Lawmakers should ask what a company losing $80 billion in six months owes the public if it fails. Anthropic wanted to be trusted with a world-changing technology. Fine. Trust is something you earn with transparency, not a ticker symbol.
How it may affect me
If you own broad index funds or a 401(k), you may end up holding Anthropic shares indirectly once it lists, depending on index rules. Check your funds' tech exposure. If you consider buying directly, read the filing's risk factors on debt and losses before the opening-day hype does your thinking for you.
