Another Crazy Day in AI: Human Rights, Machine Speech, and Everything Between

Hello, AI Enthusiasts.

Back at your desk and catching up?

A recent podcast brought together an esteemed economist and a nation’s leading legal scholar for a conversation about liberal democracies, AI, and the blurry line between human and machine. They touch on everything from everyday policies to the challenges of updating frameworks for rights and freedoms in a world of machine-assisted creation.

Meanwhile, the FDA is going big on AI for internal workflows... rolling out agentic tools to all staff. And if you think small can’t compete with big, Runway just proved otherwise with its new video-generation model. New month, same chaos, more AI.

Here's another crazy day in AI:

• Cass Sunstein on liberalism, rights, and machine speech

• FDA launches agency-wide agentic AI platform

• Runway’s new video model outperforms top tech rivals

• Some AI tools to try out

---

TODAY'S FEATURED ITEM: Rights, Technology, and Where They Intersect

Image Credit: Wowza (created with Ideogram)

What if the biggest threat to modern liberalism isn't its ideas, but the forces pushing against it, including the technologies we're building?

In a recent episode of Conversations with Tyler podcast, economist Tyler Cowen sits down with Cass Sunstein at Harvard Law School for a wide-ranging conversation that touches on some of the most pressing questions facing liberal democracies today. Sunstein, who remains one of the most frequently cited legal scholars in the world, has an unusual claim to fame this year: five books published in a single year. Among them are works examining liberalism, the capabilities and limitations of artificial intelligence, and the growing concern around manipulation in both commercial and political spheres. Their discussion moves fluidly between immigration enforcement, the constitutional status of AI-generated speech, and whether our current frameworks for rights and freedoms can adapt to a world where the line between human and machine contribution grows increasingly blurred.

Here's what came up in the conversation:

• Liberalism faces pressures that stem from long-standing human tendencies toward fear, order, and certainty.

• Immigration remains a complicated area where humane treatment, legal boundaries, and effective enforcement meet.

• Many current contributors to liberal thought come from economics and behavioral science rather than traditional philosophy.

• Certain strands of modern activism draw on classic liberal critiques but sometimes test norms of open disagreement.

• AI does not hold individual rights, though AI-generated content raises questions around responsibility and constitutional interpretation.

• Sunstein suggests considering a legal right not to be manipulated as influence becomes more targeted and data driven.

• The role of AI in future judicial processes brings up questions about fairness, trust, and the value of human judgment.

There's something refreshing about listening to two people think through hard problems without pretending they have everything figured out. Sunstein tells a story from his time at the Department of Homeland Security, when he visited the southern border and briefly spoke with two Russian men waiting in line. Just asking them if they were okay—that simple human acknowledgment—stuck with him as a reminder that policy debates always involve real people in specific situations. The conversation about AI and constitutional rights has a similar quality. Nobody's claiming to know exactly where the lines should be drawn. A toaster that says "toast is ready" obviously doesn't need First Amendment protection, but what happens when AI and human contribution become so intertwined that separating them gets messy? These remain open questions.

The discussion keeps returning to something important: liberal societies need more than just well-designed institutions and clear laws. They depend on people actually practicing cooperation, assuming good faith in others, and maintaining respect across disagreements. Those habits prove surprisingly fragile under pressure. Cowen asks pointed questions throughout—whether immigration enforcement can avoid brutality, whether libel law can still function, whether we can meaningfully distinguish manipulation from ordinary persuasion. Sunstein responds honestly, sometimes suggesting new approaches, sometimes acknowledging we're still working through the implications. For anyone trying to understand the challenges facing liberal democracies right now, this conversation offers a grounded look at where the real complications lie and why even experts are still figuring things out.

Check it out here.

Listen on Spotify here.

Watch on YouTube here.

Listen on Apple Podcasts here.

---

OTHER INTERESTING AI HIGHLIGHTS:

FDA Launches Agency-Wide Agentic AI Platform

/FDA News Release

The FDA has officially rolled out agentic AI tools to all agency employees, giving teams the ability to build more complex, multi-step workflows that support tasks across reviews, inspections, compliance, and administration. The rollout builds on the success of Elsa, the agency’s earlier LLM tool, which more than 70 percent of staff adopted voluntarily. This new system emphasizes human oversight, strong security controls, and strict protection of sensitive data, especially since no input is used for model training. The agency is also launching a two-month internal challenge to encourage staff to build practical AI solutions ahead of FDA Scientific Computing Day in early 2026.

Read more here.

---

Runway’s New Video Model Outperforms Top Tech Rivals

/Ashley Capoot, Reporter, on CNBC

Runway has unveiled Gen 4.5, its latest text-to-video model, which currently holds the top position on the independent Video Arena leaderboard—surpassing Google’s Veo 3 and OpenAI’s Sora 2 Pro. The model can generate high-definition videos based on written prompts and shows strong performance in physics, motion, and camera-aware reasoning. Runway leaders say this milestone reflects how a focused team can outpace even the largest AI companies, with Gen 4.5 serving as the first in a series of major upcoming releases. The model will roll out to all customers within the week via Runway’s platform, API, and partner integrations.

Check it out here.

---

SOME AI TOOLS TO TRY OUT:

• Ripplica – Automate browser workflow by recording it once and letting AI repeat it for you.

• Karumi – Create personalized live video demos for prospects, in any language.

• Validate Idea – Test your idea, collect feedback, and decide whether to build or pivot.

---

That’s a wrap on today’s Almost Daily craziness.

Catch us almost every day—almost! 😉

---

EXCITING NEWS:

The Another Crazy Day in AI newsletter is on LinkedIn!!!

Leveraging AI for Enhanced Content: As part of our commitment to exploring new technologies, we used AI to help curate and refine our newsletters. This enriches our content and keeps us at the forefront of digital innovation, ensuring you stay informed with the latest trends and developments.