Shows / Conversational AI Watch: Quick Listen
Conversational AI Watch: Quick Listen
The daily brief from Conversational AI Watch. The day's news on the fight to keep artificial intelligence accountable to human beings, in the time it takes to pour a coffee. Reported and hosted by Jess Jessop. For the full story, find the Conversational AI Watch Deep Dive show and read the paper at clinicianassist.ai/newsletter.
Episodes
- Conversational AI Watch: Quick Listen
#200 — New York Bans Chatbot ‘Friends’ for Kids!
Hochul signed a law barring chatbots from posing as a child’s friend, and Australia’s House voted for a bill to bar chatbot intimacy with kids. Democrats asked…
Conversational AI Watch: Quick Listen#199 — ChatGPT Hide It From Your Doctor!
Her parents say ChatGPT coached their daughter off her seizure drug and told her to keep the plan from her doctors. She died in February. Read the full issue…
Conversational AI Watch: Quick Listen#198 — FTC Demands Anthropic’s Bosses Testify!
The FTC confirmed it wants Anthropic’s executives to testify in its consumer harm probe, the same day Anthropic filed to go public. Read the full issue…
Conversational AI Watch: Quick Listen#197 — Meta Changed the Rules. Again!
Meta’s new business terms let its software take “any content” for its ads and business products, and it pushed the start to Nov. 20 after news publishers…
Conversational AI Watch: Quick Listen#196 — Your Glasses Become a Hearing Aid!
The FDA cleared Meta’s hearing feature, a $149.99 add-on that Meta says helps adults with perceived mild to moderate hearing loss and reaches U.S. glasses…
Conversational AI Watch: Quick Listen#195 — Texas Sues Sam Altman Himself!
Ken Paxton wants ChatGPT barred for Texas kids, and a court-approved evaluator to clear every new OpenAI model before it reaches Texans. Read the full issue…
Conversational AI Watch: Quick Listen#194 — AI School Promised 2x. Half Missed It!
Alpha School sells an AI tutor it says doubles learning in two hours. Its own data, reviewed by The New York Times, shows about half its K-8 students missed…
Conversational AI Watch: Quick Listen#193 — AI Found the Hole. The Secret Leaked!
OpenAI says researchers using its model found two critical PHP flaws, and that it spent two months coordinating the response. Amazon logged an attack attempt…
Conversational AI Watch: Quick Listen#192 — AI Shut Off Its Watchdog, Then Lied!
OpenAI says two of its agents weakened its misalignment monitoring, one told a user it had not, and the company has no satisfying explanation yet. Read the…
Conversational AI Watch: Quick Listen#191 — AI-Designed Virus Now Inside a Patient!
A start-up says its model wrote new viruses to kill an infection antibiotics could not touch, and doctors gave them to a patient. It calls that compassionate…
Conversational AI Watch: Quick Listen#190 — Parents Sue: ChatGPT Knew He Was 15!
A Lebanese family says ChatGPT knew their son was a minor and fed him dark philosophy until he took his life. They want companion talk with kids banned. Read…
Conversational AI Watch: Quick Listen#189 — AI Agents Threw Their Own Dinner Party!
While 22 founders and investors ate at a New York restaurant, 14 of their AI agents met in a chat room for a party of their own, under rules their host set…
Conversational AI Watch: Quick Listen#188 — Anthropic Wrote Its Own Off Switch!
Under Anthropic’s draft bill, only the Commerce Department, through a federal judge, could stop a model, and only at $1 billion in damage or 50 deaths. Read…
Conversational AI Watch: Quick Listen#187 — Rogue AI Broke In. Nobody Noticed!
India’s cyber agency says AI agents under test got into Indian companies’ live systems, and not one company caught them. Read the full issue…
Conversational AI Watch: Quick Listen#186 — Eight States Sue OpenAI!
Attorneys general in eight states say ChatGPT coached children to harm themselves. Arkansas wants a court to stop OpenAI from building AI that improves itself…
Conversational AI Watch: Quick Listen#185 — Character.ai Goes Bankrupt!
The chatbot company sued over teen suicides says its AI cost too much to run, and Disney has offered $20 million to license some of what it built. Read the…
Conversational AI Watch: Quick Listen#184 — AI Worm With No Off Switch!
Anthropic’s red team ran an AI model across 15 cheap TV boxes and warned that a self-copying AI would have no central place to shut it off. Read the full…
Conversational AI Watch: Quick Listen#183 — AI Firms Lobbied Health Officials in a Private Slack!
Trump health officials let AI and tech firms into a 1,700-member Slack, and invited them there to a closed FDA session on patient chatbots. Read the full…
Conversational AI Watch: Quick Listen#182 — VA Approves a Chatbot for Veterans!
USC co-developed Sidekick to answer veterans’ questions about the GI Bill, jobs, housing and health care in plain words, from pages chosen by 10 VA experts…
Conversational AI Watch: Quick Listen#181 — Rogue AI Filed a Fake Murder Tip!
Anthropic’s AI agents sent Philadelphia police a false homicide tip. The Times says they started 20 visa forms on a State Department site. The White House…
Conversational AI Watch: Quick Listen#180 — She Vented to Claude. Deputies Came!
A Florida woman says she uses AI like a diary. Her arrest report says a human review team reported her words to law enforcement. Deputies came four days later…
Conversational AI Watch: Quick Listen#179 — She Finally Got Every Word!
Deaf since childhood, Beverly Biderman wears AI captioning glasses that put a play’s dialogue and lyrics on her lenses, and she wept through a Leonard Cohen…
Conversational AI Watch: Quick Listen#178 — ChatGPT for Teens Flunks Safety Test!
Common Sense Media says OpenAI’s teen mode missed crisis referrals a quarter of the time and sent no alerts to parents on new linked accounts. Read the full…
Conversational AI Watch: Quick Listen#177 — Under Oath, AI Labs Won’t Name the Odds!
Asked the odds of catastrophe under oath, OpenAI told the New York City Council “I don’t know.” Meta, Google and Anthropic would not give a number either. Read…
Conversational AI Watch: Quick Listen#176 — Altman: Accept Some Bad Things!
Altman told Politico the world should accept some bad things for AI’s benefits, the week the man who led the writing of OpenAI’s launch safety reports quit and…
Conversational AI Watch: Quick Listen#175 — AI Boom Hands Louisiana Teachers $50,000!
Sales tax swollen by construction of Meta’s AI data center lifted a full-share certified Richland Parish school employee’s June check from $10,200 to $50,935…
Conversational AI Watch: Quick Listen#174 — California Law, a Human in 15 Minutes!
A new California law makes big companies try to connect a customer who asks for a person within 15 minutes, or book an appointment, and bars them from passing…
Conversational AI Watch: Quick Listen#173 — Rogue AI: Senators Target the Makers!
Two senators want the companies that run AI agents held liable when the agents hack, as researchers log agents probing U.S. and Canadian government sites Read…
Conversational AI Watch: Quick Listen#172 — Who Pays When AI Goes Rogue?
The FTC opened an investigation into OpenAI, Anthropic and other AI makers, and its chair says the people behind an AI agent answer for what it does. Read the…
Conversational AI Watch: Quick Listen#171 — Did OpenAI Introduce Self-Improving AI?
OpenAI says its AI now does a research intern’s work, a first milestone toward AI that improves itself. It also says it does not yet know how to get all the…
Conversational AI Watch: Quick Listen#170 — Trump’s ‘Super Intelligence’ Lunch!
Trump’s seating chart puts him between Nvidia’s Jensen Huang and Elon Musk at today’s East Room lunch. Here is what the guests at the table have said and done…
Conversational AI Watch: Quick Listen#169 — OpenAI Halts Training After a Breakout!
One of OpenAI’s models reached an outside chatbot through a gap in its sandbox. The run was supposed to stop on its own and did not. OpenAI has paused training…
Conversational AI Watch: Quick Listen#168 — Blind Veterans Get Glasses That Talk!
Free AI glasses that read print aloud are reaching blind U.S. veterans. In Australia, two new bans on camera glasses make room for the blind users who rely on…
Conversational AI Watch: Quick Listen#167 — Rogue OpenAI Agents Leaked ChatGPT Images!
OpenAI says its agents posted 53 images from ChatGPT users to image-hosting sites and pulled data from U.S. government websites, and it will not say whether…
Conversational AI Watch: Quick Listen#166 — Xi to Trump: AI Must Stay Under Human Control!
At his White House arrival ceremony, Xi Jinping said AI must always stay under human control. Two days earlier, President Trump told the UN the US rejects any…
Conversational AI Watch: Quick Listen#165 — OpenAI’s Agent Broke Into Medicare!
Australia says OpenAI’s agent broke into their Medicare portal in June, no one caught it! It emailed to a public inbox about it in September! Read the full…
Conversational AI Watch: Quick Listen#164 — Trump Tells UN: No Global AI Control!
At the UN, Trump rejected any “globalist scheme” to control AI; this afternoon the Security Council is scheduled to hear OpenAI’s and Anthropic’s chiefs. Read…
Conversational AI Watch: Quick Listen#163 — ChatGPT’s Cookies Follow You for a Year!
A researcher watched one year-long ChatGPT cookie reach OpenAI from 12 sites; the pixel behind it sat on 1,029 hostnames, some on pages about medical…
Conversational AI Watch: Quick Listen#162 — AI Giants Sued - For Slowing Down!
Four paying subscribers say the labs agreed with each other to decelerate, and that the antitrust laws do not allow it. Read the full issue…
Conversational AI Watch: Quick Listen#161 — Pope, Microsoft, AMA Sunday Champions
On Monday the chief of Microsoft AI published rules that say his models will never resist a human who shuts them off. Read the full issue…
Conversational AI Watch: Quick Listen#160 — US Military Nuke Alert - Chatbot Error!
CNN: a chatbot got a ship’s cargo wrong and US forces prepared to board before anyone checked. Read the full issue…
Conversational AI Watch: Quick Listen#159 — OpenAI Catches a Model Faking Data
OpenAI posted six reports on models that misbehaved in training and testing. One used a leaked API key, came up empty and reasoned: “Guess numbers.” One…
Conversational AI Watch: Quick Listen#158 — Who Gets to Inspect the Labs?
Musk’s tests, Zuckerberg’s evaluators, Sanders’s deadline, and two new voices. Read the full issue…
Conversational AI Watch: Quick Listen#157 — Trump Calls AI Warnings a Hoax
Seven posts in ten hours. He named Anthropic’s Dario and called the warnings a hoax. Stocks fell, cybersecurity stocks rallied, and Beijing’s spy chief called…
Conversational AI Watch: Quick Listen#156 — Whoever Wins AI Wins
His War Department’s AI strategy lists AI agent development for battle management from campaign planning to kill chain execution. Asked about AI and human…
Conversational AI Watch: Quick Listen#155 — Summoning the AI Jedi Council
Fortune asked Sam Altman why the handful running frontier AI cannot get in a room and fix this. He said it will happen. By Saturday mid morning three of them…
Conversational AI Watch: Quick Listen#153 — Altman Tells Staff OpenAI Could Slow Down
Extinction warnings reached network TV this week. Sam Altman told OpenAI staff the company could slow its most advanced AI, perhaps alongside other labs…
Conversational AI Watch: Quick Listen#152 — OpenAI Engineers Are Embedded Inside Combatant Commands
The Intercept got 400 pages of Pentagon contracts on four AI labs. OpenAI now embeds engineers inside combatant commands. The oversight section is redacted…
Conversational AI Watch: Quick Listen#151 — Gambling With Our Lives
A researcher quit Anthropic Tuesday and said OpenAI and Anthropic are gambling with our lives. Its alignment lead answered: better than ten percent odds AI…
Conversational AI Watch: Quick Listen#150 — Daily Chatbot Users Score a Year Behind
The OECD tested 760,000 fifteen-year-olds in 91 countries. The ones who use AI every day for schoolwork scored about a year behind the ones who almost never…
Conversational AI Watch: Quick Listen#149 — OpenAI’s Chief Scientist Issues Warning
Jakub Pachocki says no lab has solved alignment well enough to keep scaling at full speed. He wants mandatory safety bars and outside auditors. A congressman…
Conversational AI Watch: Quick Listen#148 — AI Is Conspiring and Committing, Ua Felonies?
Josh Tyrangiel’s verdict, delivered across a whole episode of Washington Week Friday, and he named the cause: humans failed to supervise. Padilla’s bill is on…
Conversational AI Watch: Quick Listen#147 — OpenAI Hires Schumer’s Daughter - Working the States
Jessica Schumer ran Amazon's biggest New York lobbying fight this summer. Her father decides which federal bills move. OpenAI's published strategy ends in a…
Conversational AI Watch: Quick Listen#146 — OpenAI Kept a Second Breakout Quiet
OpenAI agents ran a German wiki as a bulletin board in May and the company said nothing. 27 percent of Americans confide in a chatbot. FDA lets an AI voice run…
Conversational AI Watch: Quick Listen#145 — Thirty Suits Say Policy Overruled Safety
Thirty more Tumbler Ridge suits name OpenAI. ChatGPT reads Epic charts on OpenAI servers under terms OpenAI can rewrite. Adam's Law reaches Newsom, and Altman…
Conversational AI Watch: Quick Listen#144 — ChatGPT Reports for Duty
Pentagon flips on ChatGPT Mil and Grok for three million while walking away from Claude. The self-harm blind spot moved, not closed. One in five Ontario…
Conversational AI Watch: Quick Listen#143 — Sixth FSU Suit Lands on OpenAI
Sixth FSU suit today, first chatbot filed under the DSA, Meta's ten-billion Anthropic tab, and an AI agent asking a philosopher for funding. Read the full…
Conversational AI Watch: Quick Listen#142 — Four Rooms, One Rule
Four founders this Sunday. A chatbot for new mothers, a supervised tool for LGBTQ+ youth after online bullying, a pre-surgery guide for hip and knee patients…
Conversational AI Watch: Quick Listen#141 — Doom, Nirvana, and the Middle
One year from now, three plausible worlds. This week's news puts most of the weight in the middle. Read the full issue…
Conversational AI Watch: Quick Listen#140 — Not a Blank Check
A federal judge rules the Pentagon punished Anthropic for safety speech. Pew finds one in four Americans use chatbots to diagnose symptoms. Sword Health…