Author: Luiza Jarovsky Source: https://www.luizasnewsletter.com/p/lack-of-public-trust-could-kill-ai Published: February 26, 2026 (edition #276)
"Lack of Public Trust Could Kill AI" is a February 26, 2026 newsletter essay by Luiza Jarovsky arguing that, despite continued technological advancement, the AI industry faces a public trust problem rooted in unresolved legal, ethical, and philosophical issues. Jarovsky contends that in a year when the field must prove it is not a bubble, declining public trust and adoption could be fatal for some companies. The essay assembles its argument from several contemporaneous events and economic indicators.
Summary of argument
Jarovsky's central claim is that the AI industry's safety and transparency credibility has eroded at the moment its capabilities are advancing fastest, and that the resulting public wariness is rational and earned. She frames trust collapse as potentially fatal to specific companies rather than as a generalized industry-wide trust problem, a stronger formulation than the typical framing. She closes with the recommendation she attributes to the companies' own self-interest: "If those companies want greater adoption and higher profits, and if they want to survive, they should pay more attention to people."
Jarovsky also describes what she calls a new form of morality emerging in response to industry rhetoric: opposition to a radical "AI first" mentality, to AI productivity slogans, and to discourse that treats humans as expendable, inferior, or less than machines. She presents this as a substantive cultural shift rather than only consumer skepticism.
Discourse treating humans and machines as exchangeable
Jarovsky cites Sam Altman's remarks at the India AI Impact Summit in February 2026, where Altman justified AI energy use by comparing it to the food a human needs to eat across 20 years of life plus all of human evolutionary history. Jarovsky reads this reductionist framing, which treats humans and machines as exchangeable and similar in nature, as having produced widespread public discomfort, and argues that by February 2026 such ideas have crossed a line and are no longer acceptable. She connects this to the emerging morality described above.
Mission-statement walkbacks
Jarovsky documents changes to two companies' stated commitments. OpenAI changed its IRS-disclosure mission statement in November 2025: the prior statement read "OpenAI's mission is to build general-purpose AI that safely benefits humanity, unconstrained by a need to generate financial return," and the revised statement reads "OpenAI's mission is to ensure AGI benefits all humanity," removing both the word "safely" and the financial-disinterest clause. She further notes that OpenAI disbanded its Mission Alignment team, with the team's lead reassigned as "chief futurist."
For Anthropic, Jarovsky points to changes in its Responsible Scaling Policy, specifically the removal of its promise not to release AI models if it cannot ensure proper risk mitigations in advance. Jarovsky argues that these changes leave people increasingly unsure whether they can trust AI companies to make sound safety decisions, especially given the existential stakes involved.
Executive-level safety failures
Jarovsky cites a public post by the director of Meta's safety and alignment team stating that she failed to stop an AI agent from deleting her inbox. Jarovsky argues that granting an AI agent inbox access is itself imprudent for safety and privacy, and that the decision is worse coming from the safety and alignment director of one of the world's top AI developers. She notes that the post was not followed by any pledge for stronger agentic AI safety and, in her reading, served only to increase public mistrust.
Separately, Jarovsky points to a clip from the India AI Impact Summit in which Altman and Anthropic's Dario Amodei declined to hold hands when prompted by India's prime minister, writing: "If the prime minister of the world's most populous country and fifth-largest economy cannot get these two high-profile CEOs to cooperate and appear friendly on camera, how will they behave when the cameras are off?"
Gap between claims and economic results
Jarovsky assembles several indicators that AI's measured economic impact has lagged its promotion. An NBER survey found that roughly 70% of firms actively use AI but over 80% reported no impact on employment or productivity. A Goldman Sachs executive is quoted as saying AI investment spending had "basically zero" contribution to U.S. GDP growth in 2025. A Harvard Business Review piece from February 2026 argued that AI does not reduce work but intensifies it. Jarovsky also references studies finding that AI negatively affects skill formation, especially for junior employees.
Provenance
The essay appears in Luiza Jarovsky's newsletter as edition #276, dated February 26, 2026. As an essay it advances Jarovsky's own argument and is treated here as an attributed position rather than as factual evidence; the events and figures it cites are reported as Jarovsky's selection and characterization of them.
Relationships
- authored-by: Luiza Jarovsky
- related: AI Acceleration Paradox (Jarovsky's parallel critique), AI Divides (Literacy / Occupational / Ethico-Philosophical) (related framework), OpenAI, Anthropic
- contradicts: the industry "AI Action Plan" framing of public confidence in AI