Subvolts take China's AI export curbs and OpenAI's government stake talks headline this AI policy and threat roundup. Jul 8, 2026 - Developers
- Learn about new AI capabilities and potential policy shifts affecting model access and usage.
- Business
- Understand how geopolitical factors and AI advancements might impact your company's AI strategy and security.
- Everyday users
- Discover how AI is evolving, from new creative tools to potential security threats and international policy changes.
This episode covers significant geopolitical and industry developments impacting major AI labs and their government relations.
China May Lock Down Its Best AI Models
Audio streams from Jason Howell and Jeff Jarvis. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
AI Inside hosts Jason Howell and Jeff Jarvis discuss China potentially restricting access to its top AI models. They also cover OpenAI's reported offer of a stake to the US government, DeepSeek's chip development, and the first AI agent-driven ransomware attack. The episode touches on Anthropic's Claude Cowork usage and Meta's AI integration with Instagram photos.
China might stop letting other countries use its best AI. OpenAI reportedly offered the US government a piece of its company to ease political issues. This episode looks at how these big AI decisions could affect everyone.
This episode offers insight into geopolitical shifts in AI development, with China considering export controls on its advanced models. It explores OpenAI's strategic moves to navigate US political landscapes and highlights emerging threats like AI-powered cyberattacks. For AI professionals, policymakers, and business leaders, it provides a snapshot of current industry dynamics, regulatory pressures, and the evolving capabilities of AI agents.
- HostJason Howell
- HostJeff Jarvis
- Episode Title: China May Lock Down Its Best AI Models
- Show: AI Inside
- Length: 78 minutes
- China AI Model Access: Beijing weighs restricting overseas access to China's best models.
- OpenAI Stake Offer: OpenAI offered the US government a five percent stake to ease political pressure.
- AI Ransomware: First ransomware attack run entirely by an AI agent.
This week Jason Howell and Jeff Jarvis dig into a strange role reversal in AI, as Beijing weighs restricting overseas access to China's best models while DeepSeek races to build its own chip. They also get into OpenAI offering the US government a five percent stake to ease political pressure in Washington, and the first ransomware attack run entirely by an AI agent.
Also in this episode: Alibaba bans employees from Claude Code, more than 90 percent of Claude Cowork use turns out not to be coding, Meta puts your Instagram photos into AI, wealthy families trade school for AI tutoring, Nvidia's next rack slips to 2028, and xAI becomes SpaceXAI. New episodes every Wednesday at aiinside.show.
Note: Time codes subject to change depending on dynamic ad insertion by the distributor.
CHAPTERS:
0:00 - Start
0:01:41 - Beijing is looking at curbing overseas access to China's top AI models, sources say
0:08:18 - China's DeepSeek developing its own AI chip, sources say
0:15:44 - Anthropic's Claude Cowork heads to the cloud as data shows 90% of sessions aren't for coding
0:24:30 - Anthropic expanding Claude Cowork to mobile and web, details here
0:32:29 - OpenAI proposes 5% stake to Trump administration to ease Washington pressure: Report
0:37:13 - The Socialist Temptation of Sam Altman
0:41:17 - Scoop: Trump administration lifts restrictions on OpenAI's GPT 5.6
0:43:06 - JadePuffer ransomware used AI agent to automate entire attack
0:46:46 - Introducing Muse Image and Muse Video
1:00:28 - Nvidia’s next-gen AI rack system delayed to 2028 on manufacturing snags, SemiAnalysis says
1:01:12 - Nvidia has lost $1 trillion in market cap and is all the way back to where it was before the AI boom
1:01:25 - A global workspace in language models
1:03:32 - Cloudflare’s new policy pushes AI companies to pay for publishers’ content
1:06:09 - WithBrandon Thomas WithJohn Solomon WithKatelyn Lesse, Angela Jiang Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Jason Howell and Jeff Jarvis.1 h 23Is AI Really Bringing the End of Software?
50 minQualcomm Wants Its AI Chips On Every Device
49 minAnthropic's Katelyn Lesse & Angela Jiang: Building an Ecosystem, not a Walled Garden
2 h 24Questions people ask
