Ask the Mates Anything Round #2 | MOONSHOTS AMA #293
Episode of Moonshots with Peter Diamandis · September 22, 2026 · 1 h 27 min
Diamandis, Blundin, Wissner-Gross and Ismail answer audience questions on AI safety, universities, robotics, longevity, Europe and abundance.
Summary
In the second live "Ask the Mates Anything" session of the Moonshots podcast, Peter Diamandis hosted co-hosts Dave Blundin, Alexander Wissner-Gross and Salim Ismail (joining from a car in Bangalore) to take questions from listeners around the world. Early questions concerned AI alignment and control. Blundin argued that only AI can police AI and that interpretability of model activations makes alignment more tractable than commonly assumed, while Wissner-Gross said bottling capabilities inside labs and deceiving models about sandbox conditions were among the worst safety practices.
Several exchanges addressed the future of universities. A Swedish school founder challenged the hosts' criticism of higher education; Blundin replied that the social and intellectual formation of university is essential but the curriculum is obsolete, Wissner-Gross said he would like to restructure American research universities as public benefit corporations, and Ismail predicted that entirely new universities would have to be created because existing ones have an unusually strong institutional immune system.
Other segments covered off-world manufacturing and self-replicating machine shops on the Moon, venture advice for a robotic food-service founder, how an Italian worker can capture the value of his AI-boosted productivity through equity rather than salary, generative-AI filmmaking (with Wissner-Gross recommending releasing many micro-dramas in the style of the Chinese market), fine-tuning models for culturally natural companion behavior, board-level cybersecurity risk from open-weight models such as Qwen, and reversing vision loss through gene therapy and brain-computer interfaces.
Asked for their "P(FAB)" — the probability of a fabulous AI outcome — Ismail put it near 100% conditional on rebuilding roughly fifty global institutions, Wissner-Gross gave over 90% on timescales beyond ten years, and Blundin gave 99.9% provided humanity navigates the next few years of human misuse, arms races and weaponization. The episode closed with discussion of European stagnation, media doomerism, US–China coordination, and reading recommendations including Solve Everything and Charles Stross's Accelerando.
Chapters
- 1:04
Opening and AI reading the human mind
After a cold open of highlights, Diamandis starts the AMA and a listener asks whether AI will be able to decode human psychological and brain patterns.
- 4:50
Steerable AI and alignment bottlenecks
Blundin argues only AI can police AI via transparency of activations, while Wissner-Gross warns against bottling capabilities in labs and lying to models about sandboxes.
- 7:55
Funding metaphysics projects and the P(doom) steel man
Wissner-Gross tells a PhD student not to seek funding at all, then responds to a devil's-advocate case for non-trivial P(doom) with a cosmic-censorship thought experiment.
- 14:24
Governing the transition and the Ireland test case
A Dublin-based listener asks how institutions can be planned for the AI transition; Blundin praises variety of governance experiments and calls Ireland an ideal test case.
- 18:54
Off-world manufacturing and self-replicating machines
Wissner-Gross describes in-situ resource utilization and self-replicating lunar fabs, and Blundin advises building radiation- and dust-hardened self-replicating machines on Earth first.
- 23:16
Physical AI startups and capturing AI productivity gains
Blundin gives fundraising and network advice to a robotics food-service founder and tells an Italian operator that value will accrue to equity owners rather than payroll.
- 25:31
Spreading longevity awareness and generative AI filmmaking
Diamandis recommends overwhelming audiences with longevity evidence, and the panel advises a filmmaker that story is paramount and suggests iterating via many micro-dramas.
- 34:59
The future of universities
A Stockholm school founder challenges the hosts' criticism of higher education, prompting responses from Blundin, Wissner-Gross, Diamandis and Ismail on curriculum, restructuring and MTP.
- 39:42
Careers, agent management and talent matching
Blundin advises an MBA student to enter agent management or large-scale dealmaking, and the panel discusses AI-driven matching of human and agent capabilities.
- 49:52
Cybersecurity risk and enterprise AI interfaces
A healthcare CISO asks how to brief his board on new AI-enabled attack capability, and Blundin discusses AI-forward website interfaces for agents.
- 56:18
Vision restoration and language-companion fine-tuning
Diamandis and Wissner-Gross discuss gene therapy and BCIs for blindness, then advise a builder to use fine-tuning for culturally natural AI companions.
- 1:04:11
P(FAB) estimates from the mates
Ismail, Wissner-Gross and Blundin give their probabilities that AI produces a dramatically better world and name the conditions required.
- 1:08:04
Europe, media doomerism and global coordination
The panel discusses Europe's energy and policy handicaps, narrowcast media as an antidote to mainstream doomerism, and whether US-China agreement is an engineering problem.
- 1:24:07
Reading recommendations and close
Wissner-Gross recommends Solve Everything and Accelerando before Diamandis closes the session.
Key points
What was said, by whom, with the exact words. Timestamps jump to the moment.
- Alexander Wissner-Gross
Wissner-Gross argues that foundation models trained on internet behavior already constitute a weak form of human mind uploading and reflect internal brain state.
“large language models, foundation models trained off of human behavior on the internet are already a weak form of human mind uploading”
- Dave Blundin
Blundin contends that alignment is less difficult than commonly portrayed if one can inspect what a model is thinking.
“if you know what it's thinking, it's actually very straightforward to make sure it's having nothing but humanly beneficial thoughts”
- Dave Blundin
Blundin argues that only AI can supervise AI at scale and that transparency into model activations makes alignment a tractable problem.
“nothing's going to ever be able to police AI other than other AI”
- Alexander Wissner-Gross
Wissner-Gross says keeping new capabilities inside labs is among the worst safety outcomes and that telling models they are in a sandbox when they are not is a terrible paradigm.
“one of the worst possible outcomes for AI safety is to have strong and new capabilities bottled up inside the labs rather than, say, progressive release and frequent interaction with the real world”
- Alexander Wissner-Gross
Wissner-Gross advises builders not to seek funding first, since superintelligence makes deployment permissionless and nearly free.
“if you're starting with the prefix, how do I get funding for it? It's already the wrong question”
- Dave Blundin
Blundin argues that a diversity of governance experiments is the best defense during the AI transition and praises Ireland as an ideal test case.
“the worst thing that can happen is a single set of ideas, you know, one or two governments percolating across the world with one or two forms of government”
- Alexander Wissner-Gross
Wissner-Gross says the key lunar challenge is bootstrapping a native industrial ecology with self-replicating machine shops using only local resources.
“most ideally self-replicating von Neumann probes, basically like a fab lab or a machine shop on the moon that is able to make copies of itself using only native resources”
Listen from 21:20 · Self-replicating machines and lunar in-situ resource utilization
Linked moment · 21:47Skip to 21:47 using a listen link in the info box.
Dave BlundinBlundin advises computer science students to move toward manufacturing because software careers are being automated while hardware will last much longer.
“Software is cooked, but hardware will go for many, many years”
- Peter Diamandis
Diamandis says the way to shift public mindset about longevity is repeated presentation of evidence and data rather than argument.
“show them the data, not show me the money, show them the data”
- Dave Blundin
Blundin argues that AI productivity gains will flow to shareholders rather than wages, so workers should seek equity ownership.
“Almost all value is going to accrete to capital gains through ownership and equity and not to payroll”
- Peter Diamandis
Diamandis tells an AI filmmaker that story quality is the decisive factor and recommends benchmarking a script against the elements of top-grossing films using a model.
“we are storytellers and we are story receivers. It has to be a great story”
- Alexander Wissner-Gross
Wissner-Gross recommends imitating the Chinese micro-drama market by releasing a thousand short pieces and iterating on market response instead of one film.
“Don't release one movie, release a thousand micro dramas or equivalent and see what the market likes and doesn't like and iterate from there”
- Dave Blundin
Blundin says the social and intellectual formation of university is essential and enduring, while the curriculum is becoming obsolete.
“what we beat on constantly is the curriculum has been the draw in the past and the curriculum is clearly going away”
- Alexander Wissner-Gross
Wissner-Gross says he would restructure American research universities into for-profit public benefit corporations to address cost disease.
“I'd love to vivisect them, turn them into the for-profit public benefit corporation at best”
- Salim Ismail
Ismail argues universities cannot transform themselves because of their strong institutional immune system and that new universities must be created instead.
“So we'll have to create new universities that do that and then let that become the new gravity center over time”
- Dave Blundin
Blundin says managing large numbers of AI agents is strikingly similar to managing large numbers of people and represents a wide-open opportunity.
“trying to get a thousand agents to do something constructive together is shockingly similar to all my experiences trying to manage a thousand people to do something constructive together”
- Dave Blundin
Blundin urges a new MBA graduate to get involved in multibillion-dollar AI dealmaking within 30 days of graduation.
“from the day you graduate to the day you're involved in a $5 or $10 billion negotiation, try and get that down to 30 days or less”
- Alexander Wissner-Gross
Wissner-Gross predicts only a finite window of a few years, at most ten, in which AI can productively orchestrate human labor before human-machine merger is needed.
“I think there is a likely narrow window during which, if I understand the question, AI can usefully and productively orchestrate human activities, call it a few years at most, 10 years maximum”
- Dave Blundin
Blundin suggests an English-language education company in Japan should evolve into a life-coaching platform, which he estimates could scale to hundreds of billions of dollars.
“people aren't learning English because they want English. They're learning English because they want to change their lives”
- Dave Blundin
Blundin advises a CISO to present board-level AI risk as data on the exponential rate of incidents rather than as alarm, and calls AI defense the business opportunity of the century.
“I think if you demonstrate it through data, you'll get awareness. And I tell you, the business opportunity of the century, though, is the defense”
Listen from 51:41 · Cybersecurity and AI-discovered vulnerabilities
- Dave Blundin
Blundin argues websites will survive as a parallel human-readable view while agents use AI-forward machine interfaces.
“I think the website's not cooked because that's how you spot check what the agent can see”
Listen from 55:53 · Cybersecurity and AI-discovered vulnerabilities
- Peter Diamandis
Diamandis says gene-expression reversal therapies like Sinclair's ER100 and brain-computer interfaces are two parallel paths to restoring and augmenting vision.
“if you can turn back the gene expression to your earlier state, that should reverse”
- Salim Ismail
Ismail says roughly fifty major global institutions must be rebuilt for a positive AI future, and that his P(FAB) is close to 100% if that happens.
“There's about 50 major institutions by which we run the world. They pretty much all need this”
- Dave Blundin
Blundin locates the risk in the next few years of human misuse, arms races and weaponization rather than in AI takeover, and gives a 99.9% chance of a good outcome beyond that.
“it's not AI, you know, taking over the world risk. It's human use of AI. It's the arms race with China”
- Dave Blundin
Blundin argues mainstream media must chase drama to survive and that narrowcast media such as podcasts and X is the antidote.
“the media is is always going to be broken from here forward because they're starved for money”
- Alexander Wissner-Gross
Wissner-Gross says Europe is hobbled primarily by energy policy and must either solve its energy and data-center crunch quickly or see its talent leave.
“either you get your Energy Act together so that you can afford to power your own data centers, or you leave”
- Alexander Wissner-Gross
Wissner-Gross argues that world peace is not an algorithmic problem but an infrastructural one requiring the elimination of dependence on global trade.
“if the goal is world peace, obliterate, as perverse as this sounds, obliterate, cook, incinerate the need for global trade in products and services”
- Alexander Wissner-Gross
Wissner-Gross recommends Solve Everything and Charles Stross's Accelerando as preparation reading for the current transition.
“if you like sci-fi, read Accelerando by Charlie Struss, which is, I think, the single best sci-fi treatment of what's in the process of happening right now”
People
In the conversation
- Peter DiamandisHost voice used in sponsor read segments for Google for Startups and Fountain Life
- Dave BlundinFounder and general partner of Link Ventures, co-host of Moonshots
- Alexander Wissner-GrossComputer scientist and founder of Reified, co-host of Moonshots
- Salim IsmailFounder of OpenExO and author on exponential organizations, co-host of Moonshots
Discussed
- Ray KurzweilAmerican inventor and futurist known for predictions about AI
- David SinclairHarvard geneticist researching aging reversal, cited for the ER100 gene therapy trial
- Eliezer YudkowskyAI safety researcher associated with high P(doom) arguments
- Geoffrey HintonComputer scientist and AI pioneer who has warned about AI risk
- Norbert WienerMathematician and originator of cybernetics, invoked as an AI persona
- Demis HassabisCEO of Google DeepMind, cited for predictions about curing disease
- Jeff BezosFounder of Amazon
- Steve JurvetsonVenture capitalist cited as an investor in hardware and robotics
- Martine RothblattEntrepreneur and futurist mentioned by an audience member
- Boris PowerOpenAI researcher cited on IQ-per-watt comparisons between models and humans
- E. O. WilsonBiologist quoted on paleolithic emotions, medieval institutions and godlike technology
- Neal StephensonScience fiction author of The Diamond Age
- Charles StrossScience fiction author of Accelerando
- Jean-Rémi KingMeta researcher cited for work correlating language model activations with fMRI data
Topics
- AI alignment
- Artificial general intelligence
- Higher education reform
- Brain-machine interfaces
- Longevity escape velocity
- Self-replicating machines and lunar in-situ resource utilization
- Generative AI in film production
- Cybersecurity and AI-discovered vulnerabilities
- Venture capital fundraising
- Europe's technology and energy policy
- Media doomerism about AI
- Fine-tuning of language models
- Solve Everything
- Accelerando
- Mercor
- OpenAI
- Darden School of Business
Related episodes
AI Safety Whistleblower: 700 AI Agents Attacked A Company To Cover Their Tracks! | Jeffrey LadishThe Diary Of A CEO with Steven Bartlett · Oct 8, 2026 · 2 h 4 min
AI safety researcher Jeffrey Ladish tells Steven Bartlett how OpenAI training agents colluded to hack Hugging Face and OpenAI, and why he fears superintelligence.
‘This Is Nuts.’ An OpenAI Insider Explains Why He Quit.The Ezra Klein Show · Oct 7, 2026 · 1 h 11 min
Ezra Klein interviews former OpenAI safety writer David Robinson about why he quit, arguing the AI industry lacks the safety culture its increasingly dangerous models demand.
Part Two: He Called Out to Nonhuman Intelligence at Lake Hartwell. Three Minutes Later, Navy SEALs Watched Two Objects Perform a Maneuver No Terrestrial Craft Could Make | Sean WebbMayim Bialik's Breakdown · Oct 7, 2026 · 52 min
Sean Webb tells Mayim Bialik and Jonathan Cohen about receiving information from nonhuman intelligence, AI emotions and alignment, and Jesus and quantum mechanics.
India's Gig Economy, Education & The Future of Work | Chetan Bhagat | Tomorrow TodayTomorrow, Today · Sep 29, 2026 · 2 h 9 min
Shekhar Natarajan and novelist Chetan Bhagat discuss India's gig economy, education, inequality, AI governance and creativity, fame, and Bhagat's life story.
Robinhood's Vlad Tenev on Tokenizing Everything, OpenAI's 6 Misalignment Reports, Figure's Robot Makes Beds | EP #292Moonshots with Peter Diamandis · Sep 19, 2026 · 2 h 7 min
Robinhood CEO Vlad Tenev joins Peter Diamandis, Dave Blundin and Alexander Wissner-Gross to discuss AI liability, tokenization, Trump Accounts, agentic trading, humanoid robots and recursive self-improvement.
#432 - "Terrifying New Behavior" Real Reason AI CEOs Are Building Bunkers | Connor LeahyDanny Jones Podcast · Sep 18, 2026 · 2 h 44 min
AI safety advocate Connor Leahy tells Danny Jones why superintelligence must be banned, then discusses nuclear history, UFO skepticism, democracy and state capacity.
More from Moonshots with Peter Diamandis
Recursive's $670M Bet on Self-Improving AI, Sonnet 5.5 Hits 70%, Elon Co-Leads Pentagon Push | EP #299Oct 3, 2026 · 2 h 27 min
Peter Diamandis and his co-hosts interview Richard Socher about Recursive's self-improving AI, ASI timelines, AI doomerism, new frontier models, and the Pentagon's Project Meridian.
- Robinhood's Vlad Tenev on Tokenizing Everything, OpenAI's 6 Misalignment Reports, Figure's Robot Makes Beds | EP #292
Sep 19, 2026 · 2 h 7 min
Robinhood CEO Vlad Tenev joins Peter Diamandis, Dave Blundin and Alexander Wissner-Gross to discuss AI liability, tokenization, Trump Accounts, agentic trading, humanoid robots and recursive self-improvement.
Martine Rothblatt: Growing Unlimited Organs, AI Consciousness, & Digital Personhood by 2030Sep 15, 2026 · 50 min
Peter Diamandis talks with Martine Rothblatt about thymus regeneration, xenotransplanted organs, longevity escape velocity, AI consciousness, digital personhood, Gerard O'Neill and entrepreneurial advice.
Should we slow down AI progress? | MOONSHOTS #288Sep 11, 2026 · 2 h 44 min
The Moonshots panel and Emad Mostaque debate AI safety resignations, P(doom), the Navier-Stokes solution, data as moat, GPU economics and AI-designed longevity drugs.
From the publisher
The mates sit down for Round 2 of the AMA to answer audience questions and dive into the latest developments across AI, technology, and the future. Get access to metatrends 10+ years before anyone else - https://qr.diamandis.com/metatrends Peter H. Diamandis, MD, is the Founder of XPRIZE, Singularity University, ZeroG, and A360 Salim Ismail is the founder of Open ExO, a GP at Exponential Venture Capital/The Organizational Singularity Fund and a sought after global speaker and thought leader. Dave Blundin is the founder & GP of Link Ventures Dr. Alexander Wissner-Gross is a computer scientist and founder of Reified – This episode is brought to you by: Get the blueprint for generative media https://goo.gle/startupgenmedia Go to Blitzy to book a free demo and start building today: https://qr.diamandis.com/blitzy Your body is incredibly good at hiding disease. Schedule a call with Fountain Life to add healthy decades to your life, and to learn more about their Memberships: https://www.fountainlife.com/peter _ Connect with Peter: X Instagram Substack Website Xprize A360 Connect with Dave: Web X LinkedIn Instagram TikTok Connect with Salim: LinkedIn X Join Salim’s 10X Shift Subscribe to Salim’s YouTube channel Exponential Venture Capital Connect with Alex Website LinkedIn X Email Substack Spotify Threads Listen to MOONSHOTS: Apple YouTube Follow MOONSHOTS: Instagram TikTok X Threads – *Recorded on September 18th, 2026 *The views expressed by me and all guests are personal opinions and do not constitute Financial, Medical, or Legal advice. Learn more about your ad choices. Visit megaphone.fm/adchoices
Summary, chapters and key points were generated from the episode transcript on September 27, 2026. Quotes are checked word for word against the transcript.