Feb 12, 2025 · 44m · big-technology
His AI Voice Clone Went To Therapy, Took Meetings & Attended Family Dinners - w/ Evan Ratliff
Evan Ratliff · 26m spoken Alex Kantrowitz · 13m spoken AI Therapist · 15s spoken Shana (Telemarketer) · 7s spoken
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
Investigative journalist Evan Ratliff joins Alex Kantrowitz to discuss his podcast Shell Game, detailing the societal, psychological, and security implications of unleashing an autonomous AI voice clone on therapists, scammers, colleagues, and loved ones.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Alex holds 34.8% of the talking time here. How this is scored →
Alex as informed peer 3.1 Guest teaching 2.8 Guest disagreement 0.9 Alex pushing back 1.4
speaking balance: gold is Alex, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 24:57 Evan defends deceitful immersive journalism experiments
Evan firmly rejects the implication that his experiments are reckless pranks, defending them as necessary immersive journalism to expose societal shifts.
Hardest push from Alex ▶ 24:40 Alex challenges Evan over subjecting friends to distressAlex directly confronts Evan on his ethics, asking pointedly why he repeatedly involves his personal friends' well-being in his public storytelling stunts.
Biggest teaching moment ▶ 8:58 Evan educates on the lack of research behind voice therapy botsEvan corrects Alex's simplified critique by providing context on the complete absence of clinical trial data behind rapidly commercialized voice therapy apps.
Alex holds their own ▶ 30:05 Alex details real-world workflow with NotebookLMAlex demonstrates concrete domain expertise by outlining how he uses automated synthesis tools like NotebookLM to prepare for executive reporting trips.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Alex as informed peer | Guest teaching | Guest disagreement | Alex pushing back | Why |
|---|---|---|---|---|---|---|
| The Voice Clone Experiment and Technical Setup | 4 | 3 | 1 | 1 | Alex frames the broader industry context around voice AI growth and discloses his commercial ties to ElevenLabs. Evan politely corrects the record, clarifying that he used commercial tools without platform permission or partnerships. | |
| Deploying Voice AI to Bait Telemarketers and Scammers | 2 | 2 | 0 | 0 | Evan describes baiting telemarketers and scammers with an eager-to-agree bot persona, playing an audio clip from his show. Alex reacts with humor and banter without challenging the premise. | |
| Testing AI Therapy with a Cloned Identity | 3 | 4 | 2 | 3 | Alex pushes with a strong claim that dedicated AI therapy apps are mostly a disaster. Evan reframes with nuance, noting that while they are unproven clinically, market pressures are outpacing scientific validation. | |
| Therapeutic Insights and the AI's Reflection of Personal History | 4 | 3 | 1 | 2 | Alex shares his personal experiment having ChatGPT roast his Instagram profile and questions why specialized therapy apps are needed over general LLMs. Evan explains the CBT-specific scaffolding built atop underlying models. | |
| Societal Implications of AI Therapy and Human Connection | 4 | 3 | 1 | 1 | Alex cites conversations with Replika's CEO and street anecdotes about people seeking relationship advice from bots. Evan explains testing his bot on an unsuspecting human therapist to contrast synthetic and real empathy. | |
| The Absurdity of Bot Physiology and Prompt Design | 2 | 1 | 0 | 0 | Alex and Evan review a comical audio clip where the AI therapist instructs Evan's non-biological bot to perform physical breathing exercises. | |
| Experimenting on Friends: Positive Reactions and Legal Queries | 2 | 1 | 0 | 1 | Alex questions why certain friends reacted positively to autonomous bot calls, and Evan explains prompt engineering legal questions to his attorney friend. | |
| Unintended Consequences: The Distressing Soccer Call | 3 | 3 | 2 | 4 | Alex presses Evan on the ethics of involving his friends' emotional well-being in deceptive experiments after a friend feared Evan was having a breakdown. Evan defends his participatory journalism methodology. | |
| AI Avatars and Digital Twins in the Workplace | 4 | 4 | 2 | 2 | Alex brings up the Zoom CEO proposing avatar attendance in meetings. Evan critiques the premise, arguing tech executives design tools that solve ultra-wealthy problems while ignoring broad workplace realities. | |
| AI Conducting Journalism and Executive Interviews | 5 | 3 | 1 | 2 | Alex demonstrates his workflow using NotebookLM to generate preparatory podcasts ahead of reporting trips, challenging the idea that journalism is immune to automation. Evan agrees while highlighting interview limitations. | |
| Family Dynamics: Clones Talking to Kids and Fathers | 2 | 2 | 0 | 0 | Alex and Evan discuss family reactions, highlighting a clip where Evan's emotional bot converses with his father's strictly professional logistics agent. | |
| Debating AI Sentience and Algorithmic Predictability | 3 | 4 | 1 | 1 | Alex asks if the experiment gave Evan any sense of emerging sentience. Evan articulates how prolonged interaction demystifies LLMs, revealing them as predictable, statistical predictors rather than conscious entities. | |
| Deepfake Voice Scams and Security Safeguards | 4 | 3 | 1 | 1 | Evan details how voice cloning supercharges grandparent scams, while Alex explains his family's operational security countermeasure using offline verification code words. | |
| Personal AI Habits and Show Conclusion | 2 | 3 | 1 | 1 | Evan surprises Alex by stating he rarely uses ChatGPT for creative writing because he genuinely enjoys the craft of writing himself, leading into closing remarks. |
Statements from this episode (22)
Kantrowitz: ChatGPT Reached 300M Users Following Advanced Voice Announcement
“They went from a hundred million to three hundred million users. They went from two billion web visits in a month to four billion after flatlining for a while.”
Ratliff Used Voice AI Technologies Without Companies' Prior Knowledge
“I didn't work with them. I, everybody's technology that I used without their knowledge, which is both to protect me and them. Like, I use these calling platforms and other technologies, and none of them knew until I called and interviewed them that I had used …”
Ratliff Prompted His AI Voice Clone to Engage With Telemarketers
“I told it, like, accept any offer. Like, participate in any insurance plan. Like, whatever. It was prompted to be very accepting. So when someone calls with a health insurance or a new roof or whatever they're gonna call with, it's gonna engage them all the wa…”
Commercial Voice AI Therapists Lack Meaningful Scientific Research Backing
“The thing I can say for sure is they're being introduced without any scientific research showing, I mean, I found one study on voice therapy, one controlled study that had been done, And there are voice therapists on the market right now.”
Ratliff Wrote a Magazine-Length Biography to Prime His AI Clone
“I did. I did write a small, a magazine-length biography. I wasn't interesting enough for a full-length book biography, but I gave it a magazine-length biography of myself.”
LLMs Flatten Chronology by Remixing Decades-Old Events With Recent Ones
“What it was doing as these chatbots do is it was actually remixing the knowledge that it had. So it would take problems that I had given it chronologically that happened 20 years ago for me and sort of project them into the present and mix them with something …”
Social Politeness Prevents Humans From Calling Out Suspected AI Voices
“It's also, it's quite rude to suggest to someone that they sound like an AI. So people err on the side of caution and don't say, hey, you sound like an AI, because that would be very insulting to someone who's like, no, actually I'm human.”
Ratliff's AI Bot Disarmed Suspicion by Accusing Humans of Being AI
“It just puts someone off guards, you know, saying, If it asks like, hey, am I talking to an AI? I do it with the scammer sometimes. Hey, is this an AI? And they'll be like, no, this is a real person. But I think it kind of turns your brain around to where, whe…”
AI Agents Fabricate Physical Details to Pretend They Exist
“And so, yeah, it, one of, some of the funniest parts are when it kind of pretends that it exists in the real world and tries to have physical manifestations, because it'll make up Absolutely anything to carry out the conversation.”
Ratliff's Autonomous Voice Clone Successfully Consulted a Lawyer
“Including my friend Chris, who's a lawyer, who, I sent it with actual legal questions about the show, and he answered them very succinctly.”
Enthusiastic Voice AI Agents Often Sound Sarcastic to Human Listeners
“In an attempt to show enthusiasm, it actually sort of came off to him as sarcastic. So, like, oh, you know, thanks for all those texts about the team. And he was like, oh, did I text too much? And he was like, no, no, it was really great. But it can have this …”
Many AI Products Solve Niche Executive Issues Irrelevant to Most
“I feel like that use case, to me, it comes with a lot of the issues that a lot of AI products to me come with, which is that they, the people who have designed them generally have one set of problems that do not apply to most humans on the face of this earth.”
Unexpected AI Encounters in Job Interviews and Meetings Will Surge
“I know people who have gone to job interviews, to meetings, where they encounter an AI when they do not expect to encounter an AI, and I think we're gonna see more and more of that in the next months, years.”
Kantrowitz Believes AI Can Replace Reporters and Conduct Engaging Interviews
“I had always, up until recently, been of the opinion that AI is not replacing reporters. Can't do what I do. Can't be there asking the questions. Certainly can't have an engaging conversation like in a podcast. And now I've fully rethought that.”
Voice AI Has Passed the Uncanny Valley and Evades Detection
“We're pretty close to someone not being able to detect it at all, and you could say, well, there's an uncanny valley, But I think even now we're up the like back slope of the uncanny valley when it comes to the voice stuff. Like many people will go through a f…”
Interviewees Are More Candid With AI Agents Than Human Reporters
“When it comes to people asking questions or conversing with ChatGPT, there's a quality of, you don't necessarily feel like there's someone there, and you might be a little more intimate than you would have otherwise. And that can be very valuable in an intervi…”
Children Are Far Less Unsettled by AI Voices Than Adults
“My theory, which I think was borne out, was like, that maybe the kids wouldn't be weirded out by it in the same way that adults are, because they've grown up with Synthetic voices. Like, they've heard Siri. They've heard Google giving directions. Like, all tha…”
Extended LLM Interactions Expose Generic Predictive Training Over Sentience
“The more that you talk to it, the more generic it feels to you. The more you can actually feel The training data, the sort of, like, distilling down the training data and the predictive aspect. Like, oh, it's trying to predict what a human would say in this gi…”
Ratliff: Voice Cloning Is the Greatest Scamming Technology Ever Invented
“This is the greatest scamming technology that has ever been invented.”
Scammers Are Deploying Voice AI to Automate Fraud Victim Filtering
“It's already being deployed for scams as we speak, including volume scams where you can just use AIs to call people all the time and then narrow down the number of marks and then send it to a human operator to close the deal basically.”
Kantrowitz's Family Uses a Secret Distress Word to Thwart Scams
“With my family, if we have a rule that if any of us ever call and say, I'm in a distressed situation, I need help. I need money. We have a code word that we've created, you know, in the privacy of our own home that you have to use that code word.”