The Times Tech Podcast – "The Man Behind the NHS Mental Health Bot"

Date: July 31, 2025
Host: Danny Fortson (with guest host Louisa Clarence Smith)
Guest: Dr. Ross Harper, CEO & Co-founder of Limbic AI

Episode Overview

This episode dives into the rapidly evolving world of AI-powered mental health care, focusing on Limbic AI—a clinically validated therapy chatbot used in nearly half of NHS talking therapy services. Host Danny Fortson and guest host Louisa Clarence Smith explore major recent tech news before a wide-ranging interview with Dr. Ross Harper, probing the promise, practical details, regulatory hurdles, and ethical challenges of "therapy bots." The show offers insight into how AI is transforming mental health access, the risks of an “AI Wild West,” and the regulatory reckoning reshaping the sector.

Major Discussion Topics & Insights

1. Setting the Stage: Tech Headlines (05:07–09:46)

• Figma IPO Mania

  • Louisa highlights the Figma stock market float, valued at $20B, seen as a bellwether for the AI-fueled tech bubble. Discussion centers on whether this signals transformative change or speculative excess.

• UK Online Safety Act

  • Danny explores the controversial new law aiming to add age verification and guardrails to internet access, especially for minors.
  • The hosts discuss practical enforcement challenges, circumvention via VPNs, and free speech backlash.

    "Implementation is kind of difficult... People are already very quickly trying to figure out ways around this." – Danny (07:49)

  • The tension between protecting kids and policing speech is underscored by high-profile opposition and concerns about sweeping liability.

2. Introducing Limbic AI: Mission and Mechanism (10:53–15:42)

• What is Limbic AI?

  • Dr. Ross Harper explains Limbic has built the only clinically validated and credentialed AI therapy agent, designed to scale high-quality mental health care cost-effectively.

    "We envisage the final layer in the clinician staffing pyramid to be this infinitely scalable workforce of clinically validated AI assistants that work directly with patients during treatment, but under the supervision of human clinicians..." – Ross Harper (11:26)

• Patient Experience: With and Without Limbic

  • Without Limbic: long wait times, uncertain human callbacks.
  • With Limbic: instant access via call or text, empathetic AI triage, faster bookings, and initial clinical screening.
  • Now incorporating voice AI (a recent innovation); not yet fully rolled out.

    "It's going to show you empathy, it's going to show you warmth..." – Ross Harper (12:57)

  • The bot can deliver basic cognitive behavioral therapy (CBT), helping identify and address negative cognitive patterns via practical techniques.

3. AI in Clinical Context: Validation & Regulation (15:45–20:42)

• Regulatory Approval as a Medical Device

  • Limbic is certified—a key differentiator from the generic "AI wellness bots" flooding app stores.
  • Emphasis on peer-reviewed evidence (6 papers, including in Nature Medicine).

    "We're not a cowboy who ultimately is going to leave when things get real." – Ross Harper (17:54)

  • Many AI wellness apps play in a regulatory "gray area" and may face a "great reckoning" as policies mature—especially in the US, where state-level rules are emerging.

• Key Concerns: Clinical vs. Wellness Apps

  • Ross warns of dangerous conflation between clinical-grade therapy and wellness AI companions.

    "What worries me is when these two things get conflated, often intentionally..." – Ross Harper (16:39)

  • Coming crackdowns will force AI “therapists” to rebrand or exit the healthcare market, focusing strictly on claims backed by evidence.

4. The Coming AI Therapy Reckoning & Market Dynamics (20:42–27:27)

• US Market Dynamics

  • Limbic has expanded to 13 US states, aiming for all 50 by year-end.
  • Increasing scrutiny from healthcare stakeholders and shifting state policies.

    "Healthcare stakeholders themselves are setting up their own AI councils... Procurement just got more complex..." – Ross Harper (25:25)

• Direct-To-Consumer vs. Clinical Path

  • Harper sees a saturated, noisy app store with little real impact or regulatory oversight, making the clinical route more viable and sustainable.

    "If ChatGPT is thrown into all of those [therapy apps], it's hard to see how an individual solution is going to win." – Ross Harper (23:47)

5. The Human Side: AI Empathy, Access, and Patient Bias (27:27–33:48)

• Personalization and Empathy

  • Limbic doesn’t enforce a fixed name/personality, but lets users customize—removing biases and easing engagement.
  • Surprising finding: Many patients, especially from minority backgrounds, are more willing to reach out via AI than humans due to social anxiety or representation concerns.

    "By having a non-human for that first step, you're able to help hold their hand and get them into care... we found a statistically significant uplift of 15% in individuals self-referring themselves into care..." – Ross Harper (28:34)

• AI as Amplifier, Not Replacement

  • Rejects the “doctor replacement” narrative. Instead, AI amplifies clinician reach, reduces burnout, and makes better use of existing expertise.

    "I don't think it is a helpful framing to talk naively or arrogantly about AI replacing doctors. It's not realistic." – Ross Harper (30:51)

• Rapid Industry Adoption

  • Contrasts with healthcare’s slow adoption of past technologies; now, dire staffing shortages are accelerating AI’s adoption—with proper regulation.

6. Risks, Liability, and Explainability (33:48–37:30)

• Regulation and Liability

  • Growth won’t come without risks: Harper warns that unvalidated AI solutions in high-stakes settings will cause setbacks—and perhaps tragedies that spur regulatory backlash.

    "One of my biggest concerns is that we have a huge setback as an industry because an unvalidated solution is allowed to be used in a high stakes setting..." – Ross Harper (33:53)

  • Key: Transparent, explainable AI that can be audited and show protocol adherence—not just “black boxes.”

    "Performance alone is not enough... You must focus on things like explainability, compliance and help shape the way this AI will be used in practice." – Ross Harper (35:15)

• Analogy: Self-Driving Cars

  • Like autonomy in vehicles, clinical AI must maintain performance and accountability to earn and retain trust.

7. Co-hosts' Reflections: Skepticism, Optimism & Cultural Impact (37:30–42:41)

• Cautious Enthusiasm

  • Louisa: "Clearly, AI has such potential to do good things in healthcare... but when is the regulation going to come?" (37:34)
  • Danny: "There's a lot of people who would rather talk to a bot first as a way to kind of get in the door."
  • They underscore the magnitude of unmet need and how flawed existing human systems are, suggesting AI is worth a try.

• Empathy and Cultural Anxiety

  • Louisa questions: "Can a bot actually show you empathy?" (41:26)
  • They compare recent South Park depictions of human-AI relationships to the real world, acknowledging both the promise and the loneliness these tools may inadvertently foster.

Notable Quotes & Memorable Moments

• On Regulatory Reckoning:
  "There will be a great reckoning. It’s already beginning... policy will catch up and an initial expansion will lead to a sharp contraction, where a number of solutions will have to stop operating in the high stakes regulated healthcare environment..."
  – Ross Harper (17:36)

• On Patient Uptake:
  "We found that when an AI or our AI is owning the front door to care, you see a statistically significant uplift of 15% in individuals self-referring themselves into care... even bigger uplift for individuals from minority demographics."
  – Ross Harper (28:04)

• On Limiting AI Hype:
  "I don't think it is a helpful framing to talk naively or arrogantly about AI replacing doctors. It's not realistic..."
  – Ross Harper (30:51)

• On AI as Explanation Engine:
  "You must focus on things like explainability, compliance and help shape the way this AI will be used in practice..."
  – Ross Harper (35:15)

• On Market Flood and AI Therapy App Overload:
  "If ChatGPT is thrown into all of those [therapy apps], it's hard to see how an individual solution is going to win."
  – Ross Harper (23:47)

• On Empathy:
  "Potentially controversial, is that he thinks this technology can show you empathy. And it's like, can a bot actually show you empathy?"
  – Louisa Clarence Smith (41:26)

Timestamps for Key Segments

• Tech News Recap – Figma & Online Safety Act: 05:07–09:46
• Limbic AI Overview & Patient Pathways: 10:53–14:29
• Clinical Validation, Regulation Discussion: 15:45–20:42
• US Market Expansion, Reckoning Looms: 20:42–27:27
• Patient Behavior & AI Empathy: 27:27–30:33
• Doctor Replacement Debunked & Industry Adoption: 30:33–33:48
• Risks, Liability, and Explainable AI: 33:48–37:30
• Co-host Reflection & Cultural Commentary: 37:30–42:41

Episode Tone & Style

Conversational, incisive, and at times wry. The hosts mix skepticism with optimism, candidly poking at tech hype and regulatory realities while giving Ross Harper space to outline his vision and defend clinical AI. Humor surfaces in references to startup culture, South Park, and the many dubious “AI therapists” crowding app stores.

Takeaways

• AI can expand access to mental health care, especially when validated, supervised, and integrated into clinical settings.
• The rapid spread of poorly regulated wellness apps is creating a Wild West atmosphere—regulatory intervention is inevitable.
• True clinical impact and patient safety hinge on transparency, explainability, and evidence-backed use.
• Many patients—especially those anxious about human interaction or from minority backgrounds—prefer AI as a first step.
• Cultural anxieties about empathy, trust, and human connection remain thorny and unresolved, as recent media (and everyday experience) attest.

For listeners curious about the future of mental health, regulation, and AI’s true power—and limits—this episode offers an essential, nuanced guide.