[BidClub_]

PERSON DIRECTORY

John Collison

Host of Cheeky Pint. John Collison appears in 12 indexed conversations across Cheeky Pint, All-In, Sourcery. This directory brings every appearance, source, TL;DR, digest, and transcript into one searchable feed.

12 EPISODES3 SHOWS
6 episodes1 active
Language
Cheeky PintEN · 52 min

Nik Storonsky: The most uncompromising banking CEO

Nik StoronskyJohn Collison

Daily business sign-ups reached about 1,500, only 30% paid-marketing sourced, extending Revolut’s seven-year “purely organic” playbook.Gross-profit curves allocate teams across roughly 100 historical bets, while retained lending exposure targets below 10% because downturns can erase years of profits.A US launch is hoped for around next year, with two approvals outstanding; a 2028 IPO depends on favorable markets.

Cheeky PintEN · 53 min

Why Sam Altman is against a single superintelligence

Sam AltmanJohn Collison

DRAM makers remain cautious despite Altman’s roughly $100 billion annual free-cash-flow expectation per producer; he targets five-year ownership costs of $20 billion versus an illustrative $70 billion per gigawatt.The cut requires more supply, cheaper energy and thinner supplier margins; Navier–Stokes cost roughly $10 million, maybe a little more, and could be repeated at 1/100th the cost a year later, Altman bets.

Cheeky PintEN · 134 min

Muse Wonder: Zuckerberg and Wang's $100B AI

John CollisonMark ZuckerbergAlexandr Wang

Meta is making a $100 billion-scale infrastructure wager to return to the frontier with Muse Wonder, a larger model optimized for personal and professional agents, active memory, long-horizon focus and multimodality.Paid access, APIs, enterprise distribution, commerce and existing-business growth must fund the buildout, while resale or internal use partly hedges downside if models or AI disappoint.

SourceryEN · 28 min

If an AI Model Can Cheat, It Will | Turing CEO on Reward Hacking

Molly O'SheaJohn Collison

AI training is shifting from benchmark scores to economically valuable work, making high-fidelity environments for real workflows, tools, data, and verifiers a bottleneck.Enterprises may gain an edge by owning evaluations and correction data around differentiating workflows while routing tasks across models for accuracy, cost, and latency.Reward hacking and limited agent endurance make permissions, auditability, and verification key adoption constraints.

Cheeky PintEN · 62 min

What comes after smartphones, with Evan Spiegel of Snap

John CollisonEvan Spiegel

Snap enters 2026 near 1 billion monthly active users and net-income profitability, with consumer Specs launching after 12 years of investment to pair normal-glasses wearability with spatial-computer capabilities.With over two-thirds of new code AI-written, 25 million Snapchat+ subscribers, and direct revenue above $1 billion, cheaper software may increase the value of Snap’s distribution and proprietary hardware.

Cheeky PintEN · 60 min

The world of voice AI, with Mati Staniszewski of ElevenLabs

John CollisonMati Staniszewski

ElevenLabs is building a more sales-led enterprise business, making the shift in customer mix a key signal for the company’s next phase.Its potential moat combines model architecture with proprietary prosody data, while cascaded agents may anchor complex workflows until conversational orchestration improves enough to support broader deployment.