The signal room: AI Agents in Healthcare: Why the Best AI Knows When to Stop | Kaled Alhanafi
Wednesday, September 30, 2026
Transcript intelligence
What OpenRounds heard
Source-grounded signals extracted from the episode transcript. Timestamps and supporting excerpts appear when available.
The speaker reports that their organization started AI automation with referral and call center workflows, rigorously testing one task at a time and only expanding after achieving 99% reliability with a clear escalation path for the remaining 1% to human staff, ensuring user acclimatization before proceeding to subsequent tasks.
The speaker reports that their AI agents are designed to recognize clinical limitations and escalate to human staff when encountering situations requiring specialist judgment, emphasizing that knowing when to stop is more important than automation extent.
The speaker reports that their AI agents automate referral processing by extracting information from referrals, updating the EMR, and initiating patient contact via call, text, or interaction, reducing wait times from eight weeks to seconds for scheduling tasks.
Send us Fan MailAI agents in healthcare only work if they know when to stop. Basata CEO Kaled Alhanafi explains how his team builds agentic AI for hospital and practice operations, and why trust and governance matter more than the algorithm.Kaled Alhanafi left Lyft and Cruise to build a healthcare company after his family's experiences with the healthcare system. He now leads Basata, a healthcare AI startup whose agents read referrals, call and text patients, and handle scheduling. In this episode of The Signal Room, he explains how they decide what to automate and what a human has to own.In this episode:Why capacity in healthcare is trapped in administrative work, not only in a shortage of doctorsHow healthcare AI agents handle referrals, patient calls, and schedulingWhy "a really good AI is the AI that knows when to stop"Why he says 70% of AI transformation is change managementHow agentic AI in healthcare earns autonomy one task at a timeHow to be transparent with patients when they are talking to an AI agentWhy point solutions fail and holistic approaches workWhat hospital administrators should do first when adopting AIKey numbers from the conversation:10 to 30% of physician effort goes to administrative workAbout $150 billion a year is wasted on patient no-shows in the USBy 2030, about one in five Americans will be over 65Basata has raised a little under $25 million and recently completed its Series AChapters:0:00 Welcome and Beneath the Signal0:45 Why Kaled left Lyft and Cruise for healthcare1:34 A personal path into healthcare3:36 Shadowing practices and the start of Basata6:16 The healthcare capacity problem7:19 How Basata's AI agents work8:52 The AI that knows when to stop11:16 70% of AI is change management13:26 Trust as the differentiator14:14 Scaling at the speed of trust16:41 The 2030 demand problem18:36 Why point solutions fail21:28 AI patient scheduling and no-shows23:54 Would patients rather talk to a human or an AI agent?25:15 When an AI agent should
Show transcript excerpt
Send us Fan MailAI agents in healthcare only work if they know when to stop. Basata CEO Kaled Alhanafi explains how his team builds agentic AI for hospital and practice operations, and why trust and governance matter more than the algorithm.Kaled Alhanafi left Lyft and Cruise to build a healthcare company after his family's experiences with the healthcare system. He now leads Basata, a healthcare AI startup whose agents read referrals, call and text patients, and handle scheduling. In this episode of The Signal Room, he explains how they decide what to automate and what a human has to own.In this episode:Why capacity in healthcare is trapped in administrative work, not only in a shortage of doctorsHow healthcare AI agents handle referrals, patient calls, and schedulingWhy "a really good AI is the AI that knows when to stop"Why he says 70% of AI transformation is change managementHow agentic AI in healthcare earns autonomy one task at a timeHow to be transparent with patients when they are talking to an AI agentWhy point solutions fail and holistic approaches workWhat hospital administrators should do first when adopting AIKey numbers from the conversation:10 to 30% of physician effort goes to administrative workAbout $150 billion a year is wasted on patient no-shows in the USBy 2030, about one in five Americans will be over 65Basata has raised a little under $25 million and recently completed its Series AChapters:0:00 Welcome and Beneath the Signal0:45 Why Kaled left Lyft and Cruise for healthcare1:34 A personal path into healthcare3:36 Shadowing practices and the start of Basata6:16 The healthcare capacity problem7:19 How Basata's AI agents work8:52 The AI that knows when to stop11:16 70% of AI is change management13:26 Trust as the differentiator14:14 Scaling at the speed of trust16:41 The 2030 demand problem18:36 Why point solutions fail21:28 AI patient scheduling and no-shows23:54 Would patients rather talk to a human or an AI agent?25:15 When an AI agent should reach out, and when it shouldn't27:54 Earning autonomy task by task29:45 Regulators and healthcare AI governance31:47 Why start with back-office operations35:57 Where hidden capacity lives (the Lyft parallel)39:00 Advice for hospital administrators41:50 New roles and shared responsibilityAbout the guest: Kaled Alhanafi is the CEO of Basata, a healthcare AI company. "Basata" comes from the Arabic word for simplicity. He previously worked at Lyft and Cruise.Basata website: https://www.basata.ai/Kaled on LinkedIn: https://www.linkedin.com/in/kaledAbout the host: Chris Hutchins is the author of Beneath the Signal, about the human work behind trusted data and responsible AI in healthcare. The Signal Room is a podcast for healthcare AI practitioners.Listen on your favorite podcast app: Apple Podcasts: https://podcasts.apple.com/us/podcast/the-signal-room-with-chris-hutchins/id1850988366 | Spotify: https://open.spotify.com/show/3zUsHNLvnGE90gUaNULucZFull episode page and transcript: https://signalroompodcast.com/episodes/ai-agents-know-when-to-stop#HealthcareAI #AIAgents #SignalRoomGuest: Kaled AlhanafiCompany: BasataWebsite: basata.ai———🎧 Listen on Apple, Spotify and more: https://signalroompodcast.com📨 The AI Health Pulse newsletter: https://aihealthpulse.beehiiv.com/?modal=signup📘 Beneath the Signal (my book): https://www.amazon.com/Beneath-Signal-Replace-Leadership-Healthcare/dp/B0GZPNVRCX?maas=maas_adg_6A599874CB2DA12E1ADF3BD5E6DB8463_afap_abs&ref_=aa_maas&tag=maas💼 Hutchins Data Strategy: https://hutchinsdatastrategy.comSupport the showAbout The Signal Room: The Signal Room is a podcast and communications platform exploring leadership, ethics, and innovation in healthcare and artificial intelligence. Hosted by Christopher Hutchins, Founder and CEO of Hutchins Data Strategy Consultants. Leadership, ethics, and innovation, amplified.Website: https://www.hutchinsdatastrategy.com LinkedIn: https://www.linkedin.com/in/chutchins-healthcare/ YouTube: https://www.youtube.com/@ChrisHutchinsAiBook Chris to speak: https://www.chrisjhutchins.com
Daily healthcare AI brief
Get the free healthcare AI briefing
One concise briefing built from reporting, research, policy, blogs, videos, and full podcast transcripts.
Free. No account needed. Unsubscribe anytime. Already subscribed? Sign in to save and personalize.