The pope joins the resistance, Berkeley had it with legal hallucinations, Colorado goes backwards, one in five expects unrest against AI.

Welcome to a new edition of misaligned bits, the (roughly weekly) newsletter from Misaligned where we sum up recent news and research, sometimes with a lighter touch.

As usual, we will mark all non-medium links with “➚” (external link) and all possibly paywalled links with “🔒”.

Misaligned Recap

This week in Misaligned, David Dill continues to write about his frustration with AI Assistants in his article “Hey AI, Pump The Breaks”. Also, we continue our series of academic papers recommend for reading with a reading list of “Critical Views On LLMs and Health Advice”.

Catholic bits

In his encyclical “Magnificia Humanitas”, Pope Leo writes that “the use of AI is never a purely technical matter: when it enters processes that affect people’s lives, it touches on rights, opportunities, status and freedom.” He points to the harful use of AI for “manipulation of information or violations of privacy”.

He also highlights the danger of AI in automated decision-making: “Indeed, entrusting an algorithm in practice with the power to select who is worthy or not, without anyone bearing responsibility for that judgment, is to hand over the task of redefining the boundaries of human possibilities.” (➚MAGNIFICA HUMANITAS)

It is the first time a Pope has written extensively about AI, and while his points mostly reflect common concerns, it will be interesting to watch the reactions and the debate that follows.

Regulatory bits

The British Electoral Commission has called for new legal controls over misinformation from AI chatbots, after a think tank found they had made serious mistakes during the recent Scottish election. The AI assistants had made up fake scandals, invented candidates or gave wrong date for the election (➚The Guardian).

Taylor Wessing has written up a great overview about what the EU’s Omnibus Simplification Package (➚European Council) means for the AI Act, what deadlines shift and what else is changing (➚Taylor Wessing).

On May 14 Colorado adopted a new AI law (➚SB 26–189). It replaces a 2024 law and strips three significant obligations of the statute: The requirement for risk management programs, impact assessments, and requirement to prevent algorithmic discrimination. It is unclear if the new law renders SpaceX/xAI’s lawsuit against the state of Colorado moot, as the plaintiffs in the case had complaint about exactly those requirements.

Other bits to read

Baker Botts has published an overview about AI regulation in the energy sector (➚Baker Botts). Meanwhile, the Guardian has published an article about people falsely identified as a thief by British shops using Facewatch’s face recognition system (➚The Guardian).

Legal practice bits

Berkeley Law School has adopted a new policy governing students’ use of AI. The new rules forbid the use of AI for most activities associated with academic work.

From the policy: “The use of AI is prohibited for aid in conceptualizing, outlining, drafting, revising, translating, or editing any work submitted for credit. AI use is prohibited for any use for any purpose in any exam situation.“ (➚Forbes)

One in five

New research by King’s College London’s Institute for AI and Policy Institute finds that (➚King’s College)

• Seven in 10 members of the UK public are worried about the economic impacts of AI.
• Six in 10 believe it will eliminate more jobs than it creates.
• Half of the people asked think its impact will be worse than a normal recession.
• One in five think it will create civil unrest.

Palantir bits

Meanwhile in London, a £50 million Metropolitan police deal with Palantir has been blocked by the London mayor, Sadiq Khan. The Mayor’s Office for Policing and Crime, which must approve contracts of this size, has withheld approval, saying Scotland Yard had seriously engaged with only one potential supplier.

A spokesperson for the mayor said Londoners only wanted to see public money being paid to companies that “share the values of our city”. (➚The Guardian)

Meanwhile, Palantir has responded by accusing Khan of “putting politics above public safety” (➚The Guardian).

Science Bits

In their paper on “AI ethics through a decolonial lens”, Selena Nemorin and Beatrice Bonami argue that “mainstream AI ethics frameworks fail to recognise epistemic diversity and to establish dialogical relations with non-Western knowledge systems, thereby reproducing colonial hierarchies of thought and governance within technological design and deployment”.

Data sovereignty should therefore include dimensions such as:

• Indigenous empowerment to decide who is Indigenous and from whom data should be collected to inform policies;
• Data collection that is aligned with Indigenous values, interests, and priorities;
• Freedom to decide who has access to data collected appropriately; and
• Advocating for respectful data governance, storage, security, and decision-making powers

They also argue “that decolonising AI ethics requires moving beyond the universalist and procedural tendencies of Western ethical paradigms towards pluriversal, situated, and ecologically embedded forms of ethical reasoning.”

Nemorin, S., Bonami, B. AI ethics through a decolonial lens: what does AI ethics look like if we take seriously the push to decolonise it?. AI & Soc (2026).

Follow and subscribe

If you stumbled across this article on the web, subscribe to the “misaligned bits” newsletter and follow Misaligned on Medium, LinkedIn, Threads or Mastodon.

You can now also subscribe to this newsletter directly with your email.