Good morning!

It was another busy week for us at Tech Policy Press, and for everyone concerned with issues at the intersection of technology and democracy.

One of the more significant developments this week occurred in the United States Congress. Early Wednesday, at a markup hearing to consider legislative recommendations for budget reconciliation, House Energy & Commerce Committee Republicans advanced a ten-year moratorium on enforcing state legislation on artificial intelligence.

We invite commentary and analysis on this subject; even if it fails to advance in the Senate, Republicans have indicated they would like to advance such a measure. Key questions are around definitions of AI and automated decision-making technologies, the implications of a moratorium, and what it will mean for tech accountability and policy debates in the US going forward.

Our first contributor perspective on the subject came from Public Citizen’s J.B. Branch and Ilana Beller. “This would not simply be a pause in regulation,” they argue. “It’s a decade-long permission slip for corporate impunity.”

This week, it also felt like we were on Sam Altman watch:

• The possibility of a moratorium on state AI regulation was discussed at last week’s Senate hearing on AI featuring executives from OpenAI, Microsoft, Micron, and CoreWeave. At the hearing, OpenAI CEO Sam Altman said, "the cost of AI will converge to the cost of energy... the abundance of it will be limited by the abundance of energy." Dr. Sasha Luccioni says we should pause and reflect upon the inevitability of this statement, and consider alternatives.

• With a trip to the Gulf last week, Sam Altman is selling “democratic AI” in partnership with the Trump administration. But will painting itself as red, white, and blue as a can of Budweiser be good for OpenAI’s business? And what will it mean for democracy?

• This month, World (formerly Worldcoin), the Sam Altman-backed startup that collects iris scans to provide “proof of human,” officially launched in the US. Sometimes something that looks dystopian is, in fact, dystopian, write David McNeill and Emily Tucker.

Online safety, content moderation, and free expression:

• Tech leaders were given every form of special treatment to make good on the promise of creating a "digital public square" for the democratic exchange of ideas, writes Dr. Mary Anne Franks. Not only has it failed to meaningfully challenge the rise of authoritarianism in the US, it has accelerated it.

• It appears clear that Trump and Musk’s ableist rhetoric both informs and forecasts DOGE’s work, writes Tech Policy Press fellow Ariana Aboulafia. Its violations of data privacy and expansive use of AI without proper oversight have already harmed disabled people, and will continue to do so, she says.

• In the wake of the recent India-Pakistan escalation, experts are still trying to make sense of the role that the information war played. Ramsha Jahangir spoke to two experts who tirelessly navigated the deluge of disinformation during the crisis: Pratik Sinha, co-founder and editor at Alt News—one of India’s major fact-checking websites, and Asad Baig, founder of Media Matters for Democracy—a non-profit focused on media literacy and development in Pakistan.

• If Elon Musk’s direct involvement in proliferating racialized hate online goes unchecked, it will set a dangerous precedent. Aishik Saha covers key findings from a recent Center for the Study of Organized Hate report.

• In this year’s installment of the GLAAD Social Media Safety Index, TikTok leads other major social media platforms, including Facebook, Instagram, Threads, YouTube, and X. But none of the platforms are doing enough to protect LGBTQ users from hate and to preserve their free expression, according to GLAAD.

• Imagining a "Wikipedia Liberty Index," Ryan McGrady says real or threatened censorship of Wikipedia is a lagging indicator of a state's attitudes towards free press, free speech, academic freedom, and free expression in general, and a real-time indicator of active efforts at suppression.

• Child online safety laws are proliferating. Sofya Diktas, Anabel Howery, and Robyn Caplan outline common mechanisms that US state bills provide for verifying age and challenges that may emerge from them.

We had multiple pieces that consider AI chatbots, digital duplicates, and synthetic media:

• A growing number of anecdotes and reports suggest that overly flattering, affirming AI systems may be reinforcing delusional thinking, deepening social isolation, and distorting users’ grip on reality, writes CDT's Amy Winecoff. We need systems that can introduce productive friction, she says.

• To address the challenges they present, Atay Kozlovski proposes categorizing all digital duplicates into one of four types, based on two core dimensions: whether the digital duplicate represents a living or deceased person, and whether it was created with or without the subject's consent.

• Just days after Pope Francis’ death, official White House accounts published an AI-generated image of President Donald Trump dressed in papal vestments. The President’s vague dismissal and the confusion it prompted among some in the public raise questions about accountability, writes Belle Torek.

• It is important to view image authenticity not only as a safety measure but also as a driver of digital transformation and efficiency. When services, businesses, and governments can instantly verify the integrity of what they see and receive, everything works better, writes Truepic's Mounir Ibrahim.

As always, we had multiple pieces on European matters. (Speaking of which, cheers to everyone who joined the meetup in Brussels with Tech Policy Press Associate Editor Ramsha Jahangir, Contributing Editor Mark Scott, and our friends and the Center for Democracy and Technology Europe.)

• Rather than weakening GDPR to meet the demands of corporate interests, the EU should strengthen it to ensure that the digital economy evolves in ways that protect people’s rights and foster rights-respecting innovation, writes Itxaso Dominguez.

• The European Commission’s preliminary findings in an investigation into TikTok’s ad transparency resonate with longstanding concerns from researchers and civil society groups that have struggled to use TikTok’s transparency tools effectively, writes Ramsha Jahangir.

• As artificial intelligence becomes central to global power competition, Europe finds itself at a crossroads. Clear rules can build trustworthy products, protect citizens, and foster democratic and economic innovation, writes Jonah Schwope.

• In an era when export restrictions, tariffs, geoblocks, and censorship increasingly balkanize the digital and physical worlds, the EuroStack should be a tech sovereignty project with a global and communal perspective, writes Brice Bowrey.

That’s not all this week:

• Internet governance: New podcast! Milton Mueller is the author of Declaring Independence in Cyberspace: Internet Self-Governance and the End of US Control of ICANN, a history that tells us a lot about where we are today when it comes to the geopolitics and global governance of technology, and where we may be headed. I spoke to him about the book.

• Labor and automation: Diana Enriquez and Johannes Anttila explore what automated work environments teach us about supporting workers more broadly, including through policy, strategy, and design for better technology tools.

• Antitrust: Amid a blizzard of antitrust activity, many online are repeating common misconceptions about the Big Tech cases — and it’s not just the rookies. On the bright side, John M. Newman writes, most are pretty easy to correct. Here’s how.

• Digital divide: Ending Digital Equity Act programs isn’t putting America first; it’s pulling out of a critical global race and wasting taxpayer dollars in the process, argues Jessica Dine, a policy analyst at New America’s Open Technology Institute and Wireless Future.

That’s it for this week. Thanks to all of our contributors! Please forward this email to your colleagues and encourage them to subscribe. And send us your ideas!

-Justin