The Shadow AI Problem Nobody's Talking AboutClaimPeople will use AI tools only when the tools are immediately helpful to them, so organizations should lower the barriers rather than force employees to spend separate time on experimentation.23:31
Threads / Paying for it / 2026
Paying for it in 2026
11 sessions
The Shadow AI Problem Nobody's Talking About
The Coding Agent Multiverse of MadnessClaimThe coding agent gateway is intended to give developers freedom to use their preferred tools while giving enterprise administrators centralized security, procurement, observability, and cost controls.7:27
PodcastHow We Cut LLM Latency 70% With TensorRT in ProductionPushed backMaher Hanafi rejected the initial assumption that smaller GPUs and multiple models per GPU would be the most efficient configuration.12:08
What's Special About Meta's Multi-Agent SystemsClaimSpatial and temporal frame merging reduces redundant visual input, token usage, GPU memory needs, and processing cost.18:47
PodcastAgents & the $40M Bet on Multiplayer AIPushed backStanislas Polu rejects the assumption that a model-performance plateau would immediately make inference tokens nearly free because demand and limited inference capacity could preserve high margins.42:32
PodcastAutonomous Agents at Work: From OpenClaw Hype to Enterprise RealityClaimAgent auditability should cover quality, performance, safety, cost, and business impact.8:55
The Caveman Prompting ChallengePushed backThe claim that temporary launch pricing means the pricing cannot be trusted is rejected; James Barney says consumption and the value of what the tokens accomplish are the difficult parts.9:35
How a Logistics Giant Keeps AI Data Locked DownClaimJason Ward uses an LLM observability platform's AI agent to find metrics that he did not know existed and compare those metrics with AI costs.3:42
Token Maxxing Is Over and Here Is What Comes NextPushed backDemetrios Brinkmann initially interpreted the $20,000 home-screen example as a possible large upside, while Pepijn Senders meant it as a question about whether a small improvement justified the spend.7:21
PodcastPlease Stop Telling Your AI to "Search Harder"ClaimAgent Substrate is intended to suspend and resume agents while preserving their working memory and state across hibernation cycles.31:36