83% of $92B VC goes to AI, yet AI PhDs face 120 rejections

AI may be the hottest sector in venture capital—drawing 83% of Silicon Valley's $92 billion in funding—but the talent picture is far weaker than the funding headlines suggest. A former Amazon machine-learning scientist with a PhD, after a failed startup, applied to 120 jobs in two months and got zero offers. Even in AI, employers appear to want compliant execution over original research instincts, a warning sign for the field's long-term innovation capacity.

The central finding of the New York Post's October 2, 2026 report is not another aggregate layoff statistic, but a lived paradox: Silicon Valley is absorbing record levels of artificial-intelligence investment while thousands of senior technologists with elite resumes, PhDs, and six-figure salary histories cannot find work. The human accounts in the story give the numbers a face. Ex-Google employee Basem Istanbouli says his severance was anything but generous, and that without a partner he could not have remained in the Bay Area. A former Amazon machine-learning scientist with a PhD, who left to launch a startup that failed, applied for 120 jobs in two months and received no offers. One company subjected him to six rounds of interviews before rejecting him. These are not marginal workers: they are the experienced builders who until recently earned between $310,000 and $340,000 and now describe living on savings and spouses' paychecks.

“

These are not marginal workers: they are the experienced builders who until recently earned between $310,000 and $340,000 and now describe living on savings and spouses' paychecks.

The macro data make the disconnect clearer. Joint Venture Silicon Valley's 2026 Index reports that Silicon Valley attracted $92 billion of venture capital last year, with 83% of that capital going to AI firms. Yet the same region lost 13,100 jobs. Meanwhile, nationwide software job postings are 23% below their pre-pandemic level, according to Indeed. The implication is that AI investment is not producing broad technology employment. Capital is concentrating in a relatively narrow set of foundation-model, infrastructure, and research companies that require fewer, often highly specialized roles. That concentration does not absorb the displaced senior programmers, data scientists, and engineering leaders who once staffed the broader software economy. At the same time, AI itself is beginning to automate portions of the coding and data work those workers previously performed, further weakening demand for the 'workhorse' roles that once defined Silicon Valley.

The former Amazon machine-learning scientist's experience illustrates a second structural problem: the penalty attached to founder experience. He told the New York Post that 'having run a startup counts against you.' In his telling, tech employers want compliant execution—people who finish code and do what they are told—rather than employees with independent ideas. That cultural judgment has real economic consequences. If failed founders cannot re-enter salaried work, the risk-reward calculation for launching a startup worsens. Fewer people will leave safe corporate jobs to create new companies, which could reduce the very entrepreneurial activity that Silicon Valley's venture capital system depends on. The report's quote—'In tech you're a workhorse. You do what you're told. You're not supposed to have ideas'—is a warning sign about the sector's changing culture and its long-term innovation capacity.