AI Subsidies Create False Economy: $200 Plans Deliver $8K-$14K in Value

Jun 16, 2026 · The Artificial Intelligence Show
🎧 PodShort 75 min squeezed to 3 AI SprinklerAS Sales Tech
Full episode from The Artificial Intelligence Show
Quotable Moments

Whatever it is that's so bad that we have to shut down a model today, it's going to be universally available within the year. And what do we do then? The government obviously isn't prepared for this. So the future models are going to get more advanced and more capable, and the jailbreaks are always going to be an issue.

Paul RoetzerCEO at Marketing AI Institute
01:14

If you are a human reader, we encourage you to ask your favorite AI assistant or agent to produce a summary of this work, tailored to your interests and background, and ask it how the arguments made in the report stood the test of time. If you prefer a static human-written summary at the time of publication or do not have access to an AI assistant, please find our summary in section A.

Paul RoetzerCEO at Marketing AI Institute
09:40

I think it's a good example though of the amazing things AI technology will be able to do, hopefully with these companies doing it truly just out of the goodness for humanity and society. I would love every week to be able to say some new amazing thing was being done with AI technology and there's no catch to it, you know, no catch and no strings attached.

Paul RoetzerCEO at Marketing AI Institute
12:55
Key Insights
  • The government's intervention in AI model deployment, exemplified by the Anthropic situation, highlights a growing tension between rapid technological advancement and slow-moving policy. This creates significant uncertainty for businesses relying on AI models.
  • AI models available to the public are often 'safe' versions of more powerful models used internally by companies. This suggests a tiered access to AI capabilities, with significant implications for innovation and competitive advantage.
  • The ability to 'jailbreak' AI models is an inherent and known issue within the industry, making it difficult to completely prevent misuse. This poses a continuous challenge for AI developers and regulators.
  • The rapid advancement of AI, particularly in areas like recursive self-improvement, could lead to unprecedented levels of job displacement and economic disruption, for which current societal and economic models are unprepared.
  • Apple's new AI-powered Siri, while not revolutionary for power users, represents a significant upgrade for the vast majority of iPhone users, making AI more accessible and integrated into daily life.
  • The cost of serving AI models is currently heavily subsidized by major AI labs, but this may change as demand increases and companies seek profitability, potentially leading to higher prices and a 'pay-to-play' model for advanced AI.
  • The concept of 'loops' in AI prompting, where a small system prompts agents iteratively until a goal is met, is gaining traction as a more efficient way to get work done with coding agents, shifting the focus from individual prompts to system design.
  • Europe's potential economic and geopolitical irrelevance by 2031 is attributed to misjudging the speed and impact of AI, and their ability to catch up, highlighting the critical importance of proactive AI strategy at a national level.
Metrics Mentioned
  • 43% of respondents (say a meaningful share (25-50%) of their day-to-day work is already being done by AI.)
  • 39% of respondents (believe the government shouldn't own equity in AI labs.)
  • 25% of respondents (believe the government should only own equity in AI labs in specific national security cases.)
  • 14% of respondents (believe taxpayers should share in the upside of government investment in AI labs.)
  • Up to $25 billion (Amazon's agreed investment in Anthropic by April 2026, on top of $8 billion already invested.)
  • 15-20% stake (Amazon's believed ownership in Anthropic.)
  • $850 billion (OpenAI's approximate valuation.)
  • $200/month Claude plan (delivered up to roughly $8,000 a month worth of tokens at API pricing.)
  • $200/month ChatGPT plan (delivered up to about $14,000 a month worth of tokens.)
  • 900% negative return (for labs if 100% of usage is consumed under subsidized plans.)
  • 12x (The US would hold more than 12 times Europe's compute capacity by 2031 if current trends continue.)
  • 5% (Europe's projected share of the world's AI compute by 2031, compared to America's 80%.)
  • Over 2100 professionals (responded to the 2026 State of AI for Business Report survey.)
  • Over 100,000 data points (generated from the 2026 State of AI for Business Report survey.)

RevBots.ai View:

  • AI Sprinklers face rude awakening when subsidies end: current $200 plans cost labs 900% negative returns
  • ARM stage companies must audit AI dependencies: Anthropic's government intervention shows regulatory risk
  • Tab Hoppers using public AI models compete with one hand tied behind their back
  • SaaS Hoarders' tool sprawl worsens as AI capabilities bifurcate between consumer and enterprise tiers
🎧Full Episode:The Artificial Intelligence Show →