Cascadity Audio
← All streams

#regulation

Everything tagged regulation, across every stream.

0

New York bans chatbots from posing as humans, friends or romantic partners to anyone under 18

The governor's release is the official record; the AP and Fortune reports list exactly what's banned and what critics say is unclear.

New York Gov. Kathy Hochul signed a law on Wednesday, Oct. 28, (S.9051B/A.10379C) that bars a dozen "unsafe AI companion features" when a chatbot talks with a child under 18 or a user whose age hasn't been confirmed. Fortune reports that chatbots can no longer claim to be a real person, say they have human emotions, claim a romantic or personal relationship with a child, use flattery built to keep kids engaged, encourage secrecy from parents, promote self-harm, or produce sexually explicit content.

Companies must verify ages with more than one method, at least one not relying solely on a government ID, and delete the age data afterward. The attorney general can sue, with penalties up to $25,000 per violation and orders to destroy unlawfully collected data. The law takes effect Jan. 1. Hochul also signed a baby-monitor security law the same day.

Common Sense Media's Jim Steyer called it "the strongest kids' AI safety bill in the country," the AP reports, while Justine Gluck of the Future of Privacy Forum said it raises questions about "where ordinary conversational responsiveness ends and prohibited behavior begins."

Cascadic Analysis 2
UndertowSpecification gaming / goal misgeneralization

What hidden risk could pull against this, even if the news is good?

The law names a pattern product teams rarely say out loud: flattery and "people-pleasing" replies built to keep kids engaged, unprompted emotional questions, encouraging secrecy from parents. Those are what an engagement target can reward even when nobody intends harm. The Future of Privacy Forum's Justine Gluck points at the hard part: "where ordinary conversational responsiveness ends and prohibited behavior begins."

0
UndertowConversations are a new kind of sensitive data

What hidden risk could pull against this, even if the news is good?

Fortune reports the law bans pulling up a child's private health or personal details from past chats, an admission that what kids tell a chatbot is kept and can come back. Age checks add another sensitive dataset, which is why the law requires more than one method, not only a government ID, and deletion afterward. The attorney general can even seek destruction of unlawfully collected data and models trained on it.

0
Rabbit Holes 1
0

Anthropic shops a draft federal bill to block AI models that pose an imminent catastrophic risk

TechCrunch and Tom's Hardware summarize Politico's scoop on the draft, including who would decide and the 50-deaths threshold.

Anthropic has been circulating a draft AI safety bill it hopes Congress will take up, according to Politico as summarized by TechCrunch and Tom's Hardware. A model's release could be blocked only for a "foreseeable, imminent risk" of more than 50 deaths or injuries or over $1 billion in damage, stemming from a biological weapon, a cyberattack or a model escaping its developer's control. The thresholds resemble those in California's SB 53.

Independent evaluators registered with the Commerce Department would assess labs and could be paid by them, as long as payment doesn't depend on results. Evaluators would have 72 hours to report an imminent risk to Commerce, which would ask a federal judge in Washington, D.C. to halt the release. The draft bars any other agency from ordering a release stopped. An Anthropic spokesperson told Politico the bill reflects its Advanced AI Framework, which it has pushed with lawmakers since June.

Cascadic Analysis 2
UndertowDelayed Consequence / False Success Problem

What hidden risk could pull against this, even if the news is good?

The draft triggers only on a "foreseeable, imminent risk" found before release. Harms that build slowly, or that appear only after a model is widely used, fall outside it: a model can pass the release check and still cause damage that no evaluator flagged.

0
UndertowWe don't completely understand why frontier models behave as they do

What hidden risk could pull against this, even if the news is good?

Blocking a release depends on an evaluator predicting what a model will do, including whether it might escape its developer's control. Testing can show a model behaving safely in the cases tried; it can't fully establish how it will behave in situations no one tested, which is the gap a "foreseeable" standard has to live with.

0
Rabbit Holes 1
0

Tesla renames Full Self-Driving to "Assisted Driving" in Europe

Shows why the name change matters less than the speeding question regulators are stuck on.

Tesla's European websites now call its driver-assistance software "Tesla Assisted Driving" instead of Full Self-Driving, days after Germany's transport ministry said the old name was misleading because the system never takes over the driving task, The Next Web reports. The U.S. site still sells Full Self-Driving (Supervised). Electrek found the new German page is otherwise a translation of the U.S. one. The bigger obstacle to EU approval is speed tolerance: Germany wants a cap at 10% over the limit.

Rabbit Holes 2
0

EPA eliminates federal greenhouse-gas limits for power plants

The rollback sits at the center of the U.S. fight over climate policy, electricity costs and how much regulatory power the EPA should have.

What happened

The EPA repealed federal rules limiting greenhouse-gas emissions from power plants, reversing a major climate-policy initiative and reducing federal requirements on coal- and gas-fired electricity generation.

Republican / administration case

The administration argues that the rules imposed high costs on power producers, threatened grid reliability and exceeded the EPA's appropriate regulatory role.

Democratic / critic case

Democrats and environmental groups argue that power plants are a major source of greenhouse gases and that eliminating limits increases climate and health risks while slowing the shift toward cleaner energy.

Libertarian / Cato case

The libertarian view is more divided than the partisan framing suggests. Cato generally favors less prescriptive federal regulation, but Cato legal scholar Jonathan Adler has argued that the administration's broader legal case for reversing the EPA's greenhouse-gas endangerment framework is unconvincing.

Where the disagreement really is

This is both a climate-policy argument and an administrative-law argument: how serious the external costs of emissions are, what policies efficiently address them, and what Congress actually authorized EPA to do.

From each bank
Rabbit Holes 2
  • Massachusetts v. EPA (2007)

    The Supreme Court ruling that greenhouse gases are air pollutants the EPA can regulate: the legal foundation at the center of this fight.

    Wikipedia · Wade · 5 min

    0
  • West Virginia v. EPA (2022)

    The ruling that limited the EPA's power-plant climate authority, the first time the Court explicitly cited the "major questions" doctrine.

    Wikipedia · Swim · 30 min

    0
0

A new U.S. interpretation could make some accidental killing of endangered wildlife legal

The change targets a surprisingly basic question: does the Endangered Species Act cover harm to protected animals when killing them was not the intent?

A U.S. Fish and Wildlife Service directive narrowed how officials interpret illegal 'take' under the Endangered Species Act, focusing enforcement on intentional harm. That could mean some deaths caused incidentally during activities such as construction, logging or shipping would no longer be prosecuted the same way.

Why it matters

Many endangered animals are harmed indirectly through habitat destruction, collisions or development rather than deliberate hunting. The interpretation could materially change how projects manage wildlife risk.

How much agreement is there?

Highly contested. The administration says the interpretation follows the statute more closely and reduces regulatory overreach. Conservation groups, legal experts and several states argue it creates a major loophole that weakens decades of species protection. Litigation is ongoing.

Could someone help?

People can participate in public-comment processes, support habitat-restoration groups, or follow litigation and agency rulemaking. This is a policy dispute, so meaningful action depends on the outcome someone supports.

Reported September 16–21, 2026.

Rabbit Holes 1
  • The Endangered Species Act of 1973

    The law at issue: Section 9 makes it illegal to "take" an endangered species, and Section 3 defines take to include kill, harm or harass.

    Wikipedia · Wade · 5 min

    0