AlphaMaven AlphaMaven

Training AI on ‘human values’ can’t give an open society the dissent it needs: Harcourt

Reported by Top1000funds
Nov 20, 2026· 2 min read
Save Post LinkedIn Email
Key Takeaways
  • • Bernard Harcourt, a Columbia University law and political science scholar, argues that training AI systems on "human values" cannot supply the dissent an open society requires, according to Top1000funds.com.
  • • The commentary frames AI alignment to a single set of human values as a risk to pluralism, since it can narrow the range of acceptable views rather than sustain democratic disagreement.
  • • The piece is an opinion/commentary with no transaction, fund, or regulatory event, so its relevance to institutional investors is indirect, centered on governance and technology risk.
Summary
Top1000funds.com published a commentary by Bernard Harcourt arguing that efforts to train artificial intelligence on "human values" cannot provide an open society with the dissent it needs. The article's source text available to this summary consists largely of the publication's boilerplate description, so the specific arguments, examples and evidence Harcourt offers could not be fully verified here.

The headline frames the central thesis: that value-alignment approaches to AI, which seek to encode a shared set of human values into models, risk homogenising the views that AI systems reinforce or permit. Harcourt's position, as reflected in the headline, is that dissent, meaning organised and legitimate challenge to prevailing norms, is a structural requirement of open societies and that AI trained toward a consensus of values may undermine it.

For the institutional investment audience that Top1000funds.com serves, the piece sits at the intersection of technology governance and social license. Asset owners and managers increasingly face questions about how AI tools are deployed in decision-making, research and stewardship, and commentary of this kind contributes to the debate over whether AI governance frameworks should prioritise pluralism alongside safety. The article does not report any transaction, fund launch, regulatory action or market development.

Go deeper than the headline

Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.

Start your free 7-day trial

Sources

1 outlet

AlphaMaven summarised this story from the reporting below. Each link opens the original article.

Top1000funds Training AI on ‘human values’ can’t give an open society the dissent it needs: Harcourt