What Does the GPT-5 Update Mean for the AI Race? WTF With Emad Mostaque, Alexander Wissner-Gross, Dave Blundin, and Salim Ismail

TL;DR

GPT-5 advances the AI race mainly by cutting AI costs at least in half, matching competitors in coding, and giving 700 million people access to state-of-the-art AI. The WTF panel considered the release a meaningful step toward AGI rather than AGI itself or a tenfold leap over earlier models. Read on for their assessment of its routing system, market reaction, launch presentation, and potential business impact.

Transcript

a huge amount of expectation on GPT5. The anticipation of this launch was up there with the top three product launches of all time. This is when you see real big things happening. Either a productivity boom or the inverse. You can see you're just reaching that level now across just about everything. They cut the cost of AI at least in half if not m... Read More

Key Insights

• GPT-5 cut the cost of AI at least in half if not more, which panelists argue is its most consequential change because it makes frontier-level intelligence far more accessible to a much larger base of users and application builders.
• GPT-5's real net impact is lifting hundreds of millions of users up from a model like GPT-4o to a frontier model, which Alexander Wissner-Gross argues is as economically relevant as any single jaw-dropping new qualitative capability.
• GPT-5 caught up to everybody else in coding, described as a big deal, though live coding demos triggered a market dip and then a huge plunge in betting-market confidence about OpenAI's lead.
• GPT-5 is structured as a multi-routing system spanning from mini up to pro, described by Emad Mostaque as basically O4 but with one front layer, delivering a single interface for quick answers or long detailed research and coding.
• Sam Altman said GPT-5 is a significant step on the way to AGI, which also means it isn't AGI yet, framing the launch as progress rather than the arrival of artificial general intelligence.
• Polymarket sentiment on which company has the best AI model inverted during the launch, moving from about an 80% chance OpenAI would lead to bettors favoring Google for both the month and the year.
• OpenAI's launch felt folksy and startup-like, resembling a high school presentation, contrasting sharply with Google IO's high showbiz value and heavy use of computer-generated video.
• Taking 700 million people and suddenly giving them access to state-of-the-art AI is framed as transformative, the scale that makes the launch matter even without shocking new capabilities.

Questions & Answers

Q: What does the GPT-5 launch mean for the AI race?

The panel said GPT-5 cut AI costs at least in half and caught up with competitors in coding. They viewed its greatest impact as bringing state-of-the-art AI to 700 million people, even though the model was not considered ten times better than what came before.

Q: How much did GPT-5 reduce the cost of AI?

According to the panel, GPT-5 cut the cost of AI at least in half, if not more. They described this reduction as consequential because it makes frontier-level AI more accessible to users and application builders.

Q: Is GPT-5 considered AGI?

No. Sam Altman described GPT-5 as a significant step on the way to AGI, which the panel noted also means it is not AGI yet.

Q: What did the panel say about GPT-5's coding capabilities?

The panel said GPT-5 caught up with everybody else in coding and called that a big deal. However, Polymarket confidence in OpenAI dipped during the first coding demonstration and plunged during the second.

Q: How does GPT-5's multi-routing system work?

Emad Mostaque described GPT-5 as a multi-routing system spanning from mini to pro and as basically O4 with one front layer. The system provides one interface for quick answers as well as longer research and coding tasks.

Q: Why did Polymarket bettors shift from OpenAI toward Google?

OpenAI entered the launch with roughly an 80% chance of leading on Polymarket, but confidence fell during the coding demonstrations. Sentiment ultimately inverted, with bettors favoring Google to have the best AI model at both the end of the month and the end of the year.

Q: Why did the panel consider GPT-5 transformative despite an incremental launch?

The panel focused on scale rather than a single shocking capability. They argued that suddenly giving 700 million people access to state-of-the-art AI, and moving hundreds of millions from a model like GPT-4o to a frontier model, could be economically transformative.

Q: How did OpenAI's GPT-5 presentation compare with Google IO?

The panel characterized OpenAI's launch as folksy, startup-like, and similar to a high school presentation. They contrasted it with Google IO's polished showmanship and extensive use of computer-generated video, concluding that OpenAI's presentation underdelivered against the enormous anticipation.

Summary & Key Takeaways

• Peter Diamandis hosts a WTF podcast episode with Salim Ismail, Dave Blundin, Emad Mostaque, and Alexander Wissner-Gross to break down the GPT-5 launch. Expectations were among the top three product launches of all time, but the panel found the model solid rather than ten times better than what came before.

• The guests agree GPT-5's cost reduction, at least half if not more, and its coding gains matter most because they lift hundreds of millions of users to a frontier model and enable new applications, agents, and GPTs to be built on top over the coming months.

• Polymarket data showed confidence flipping from OpenAI to Google during the launch, with dips during coding demos. Panelists criticized OpenAI's folksy, high-school-style presentation versus Google IO's polished, video-heavy showbiz, suggesting the event underdelivered on its enormous anticipation.