Follow the Money Friday: The $100 Million AI Influence Play That Fizzled
AI-generated image · US National Wire
Leading the Future attempted to buy a regulatory shield for Silicon Valley, but the P&L on their midterm gamble suggests a masterclass in inefficient scaling.
In the venture world, we talk about 'blitzscaling' as a way to capture a market. But in the political arena, spending aggressively without a clear path to ROI is just a bonfire of capital. Enter Leading the Future, a pro-AI super PAC network that, as Wired first reported, entered the 2026 midterm cycle with a $100 million war chest and a mission to oppose policies that stifle innovation.
**Opinion:** From where I sit, this wasn't a strategic investment; it was a political gamble that yielded zero meaningful return. When you spend millions to 'bully' lawmakers and end up with a roster of safe-seat winners and high-profile losses in the races that actually mattered, you haven't scaled influence—you've just subsidized the status quo.
According to reporting from Wired, the network was funded by some of the biggest names in the Valley, including OpenAI’s Greg Brockman, Palantir cofounder Joe Lonsdale, and Andreessen Horowitz. The structure was designed for 'maximum flexibility,' utilizing a network of 501(c)(4)s and super PACs: American Mission (targeting Republicans) and Think Big (targeting Democrats).
However, the execution was a mess of inefficiency. Wired reports that Leading the Future spent roughly $26 million on primaries alone, including a costly attempt to bury Democrat Alex Bores in New York’s 12th congressional district. Bores, a former Palantir engineer, ran on an AI safety platform and was backed by the pro-AI safety network Public First (funded by Anthropic), which spent $15 million. While Bores didn't win, he finished second to Micah Lasher, who also campaigned on AI safety and tech accountability. The attempt to 'put his head on a spike' instead elevated the safety debate.
As the general election approached, the network's aggression evaporated. Wired notes that funding dried up, with zero donations reported in the second and third quarters. The spending shifted toward 'safe' candidates—an effort Sacha Haworth, executive director of the Tech Oversight Project, describes as an 'influence play' to ingratiate the network with incoming members of Congress rather than shifting the needle on legislation.
When the network actually took risks, it failed. American Mission poured over $2 million into South Carolina's Darline Graham (dubbed 'Data Center Darline'), who lost her race with 46.5% of the vote. In North Carolina’s 1st district, American Mission backed Republican Laurie Buckhout, who lost to Democrat Don Davis. In Montana, the network backed Aaron Flint, who conceded to Democrat Sam Forstag.
The only notable win in a toss-up race mentioned by Wired was Democratic representative Jared Moskowitz in Florida, funded via Think Big, who defeated Scott Singer.
Leading the Future ends the cycle with just over $20 million left in the bank. For a network that aimed to intimidate lawmakers into ignoring AI safety, the result was a crushing lack of impact. As Brad Carson, co-founder of Public First Action, told Wired, the midterms were a victory for AI safety and a 'crushing blow' to the industry's efforts to buy a compliant Congress.