Trump's 'Very Negative Forces' Line Just Cleared the Path to OpenAI's 2027 IPO and Anthropic's $2T Listing

The setting was a golf-club driveway in County Clare, not a podium. Asked by reporters whether the AI industry should slow development or face new regulation, President Trump answered: \"We could put guardrails. We can do this and that. But I think you have a lot of very negative forces that are bringing it up that shouldn't be bringing it up and they're bringing up things that won't happen.\" He added that the U.S. is \"leading China in AI\" and that \"whoever wins AI wins\" — a phrasing that reframes safety advocacy as a China-aiding brake on American power.

The Ireland comment was not a spontaneous reaction. It was the third leg of a stance the administration has been building since the December 2025 executive order on federal preemption and the June 2, 2026 EO 14409 (\"Promoting Advanced Artificial Intelligence Innovation and Security\"), which created a voluntary framework letting AI developers give the federal government up to 30 days of early access to covered frontier models before wider release. The legal architecture was already in place; Sunday supplied the political rhetoric.

Sam Altman told Fortune on September 12 that an OpenAI IPO in 2026 would come at an \"ill-advised moment,\" confirming a push to 2027. That same week the company crossed a $40B annualized revenue run rate, up roughly 2× from the end of 2025, driven by enterprise contracts, ChatGPT subscriptions, and a new ads business that hit a $1B ARR in August. The delay lets OpenAI reach a $1T-plus valuation while a sympathetic administration is still in power — a window that closes if a single Mythos-class incident hands Democrats a 2027 campaign issue.

Through 2025 and early 2026, an investable thesis built up around companies positioned to monetize mandatory AI safety work — third-party evaluators (METR, Apollo Research), AI-native security vendors absorbing agentic workloads, and data-governance platforms selling traceability. The Feb 23, 2026 \"AI scare trade\" showed how brittle that thesis is when regulatory momentum stalls: CrowdStrike, Datadog, and Zscaler fell roughly 11% on news that an Anthropic tool could disrupt their workloads, while Fortinet and Okta dropped about 6%. Sunday's rhetoric shifts the center of gravity in that trade from regulation-as-tailwind to deregulation-as-headwind.

If anything, the comment reinforces the upstream capex super-cycle. Nvidia just printed $96.2B of Q2 FY27 revenue (+106% YoY); TSMC raised its 2026 capex ceiling from $52–56B to $60–64B in July; Micron signed a multi-year HBM supply and co-design agreement with Anthropic on June 22 that bundled memory supply with joint engineering work. A White House that paints safety voices as anti-American is a White House that won't slow frontier training runs, which is the only variable that would soften the chip pipeline. SK Hynix and Samsung Electronics — both warning of HBM shortages lasting into 2027 — sit at the same inflection.

The investable takeaway is asymmetric. Long the upstream chip pipeline — NVIDIA, TSMC, Micron, SK Hynix, Samsung Electronics — because faster deployment is now the political default. Long the AI cloud infra landlords — Microsoft, Oracle, Amazon, CoreWeave, Nebius, IREN — because the same posture that shields labs from safety mandates also defends their compute commitments ($250B from Microsoft to OpenAI alone). And short, or underweight, the regulatory-tailwind trade in CrowdStrike, Zscaler, Snowflake, and Palantir until a high-profile agentic incident forces the administration to choose between \"winning AI\" and \"negative forces\" — a choice they have now publicly committed to not making.