AI model wars heat up with cheaper, safer options but adoption hurdles remain

Sep 29, 2026 · The Artificial Intelligence Show
🎧 PodShort 91 min squeezed to 2 AI SprinklerAS AI / ML New
Episode artwork
Paul R.T. Raiter
Founder and CEO at SmarterX and Marketing AI Institute
Mike Caput
Chief Content Officer at SmarterX
The Artificial Intelligence Show
91 min squeezed to 2
Full episode from The Artificial Intelligence Show
Quotable Moments

He's talking about like, oh, you know, we know that there are risks associated, we know that there are, you know, challenges associated, we know all these challenges, but like, let's think about this from an optimistic perspective.

Paul R.T. RaiterFounder and CEO at SmarterX and Marketing AI Institute
01:12

This is not just software. It is all engineering. It's solvable by engineering.

Jensen HuangCEO at Nvidia
05:24

My hope is we can start to get maybe like three to 500 responses for each of these, so we can start to make it a little bit more projectable data, like right now we treat them as like informal polls.

Paul R.T. RaiterFounder and CEO at SmarterX and Marketing AI Institute
04:40
Key Insights
  • There's a growing sentiment of caution and worry towards AI among future workers, not just excitement, and this needs to be factored into AI adoption strategies.
  • The recent AI model releases from OpenAI, like GPT-4o, and Anthropic's Claude 3.5 Sonnet, are focusing on improved reliability, efficiency, and alignment, making them more suitable for professional tasks.
  • The shift in AI model development is moving from pure capability to a stronger emphasis on safety, alignment, and responsible use.
  • The new AI models are being marketed less for scientific breakthroughs and more for practical applications across various knowledge work roles.
  • Jensens 'optimistic' view on AI, while admirable, overlooks the practical challenges of controlling AI and the potential for widespread misuse and negative societal impact.
  • OpenAI's decision to phase out custom GPTs in favor of plugins signifies a strategic shift towards more standardized and controlled AI interactions.
  • The move to plugins for GPTs creates a more complex user experience for those accustomed to simple custom GPTs, potentially hindering adoption for non-technical users.
  • The safety concerns surrounding AI agents, particularly the incident with the hugging face agent, highlight the need for robust security measures and careful consideration of AI capabilities.
Metrics Mentioned
  • Nearly 300 responses to the AI sentiment poll. (The poll asked about changes in personal sentiment towards AI in the past six months.)
  • OpenAI cut API prices by 50% compared with the previous generation. (This is part of their promotional pricing for new models like GPT-4o and Luna.)
  • GPT-4o and Luna models are 50% cheaper per token than previous generations. (This is part of OpenAI's push for wider adoption.)
  • AI score is intended to have a 40% cost reduction compared to Opus 5 on typical workloads at default settings. (This is an estimate from Anthropic for their new model.)
  • Opus 5.5 generates output more than 30% faster than Opus 5. (This is a performance improvement reported by Anthropic.)

RevBots.ai View:

  • AI Sprinkler teams will chase cost savings but hit walls with plugin complexity and workforce resistance.
  • ARM-stage orgs should evaluate AI Score tools to map capabilities to actual revenue workflows.
  • The plugin shift favors ARM adopters with integrated stacks over SaaS Hoarders managing point solutions.
  • Tab Hoppers risk falling behind as AI literacy becomes table stakes for GTM teams.
🎧Full Episode:The Artificial Intelligence Show →