Everyday AI Podcast – An AI and ChatGPT Podcast
Everyday AI Podcast – An AI and ChatGPT Podcast

Ep 809: OpenAI’s New GPT-5.6 Release: What's New, Why the Feds Are Blocking it and What's Next

June 30, 2026

AI Summary

5 min read

OpenAI announced GPT 5.6 on Friday, but almost nobody can use it. For the first time, a major GPT upgrade arrived as a blog post and a limited preview for roughly 20 trusted organizations, not as a Chat-GPT toggle for a billion users. The reason is not technical—it is political. The U.S. government asked OpenAI to hold back its most capable model until safety reviewers get a look, and OpenAI agreed. This episode explains what GPT 5.6 actually is, why the feds intervened, and what the new permission-slip era means for businesses that depend on frontier AI.

What OpenAI Actually Built

GPT 5.6 is not a single model but a three-tier family named Sol, Terra, and Luna—Sun, Earth, and Moon. Sol is the flagship: built for the hardest reasoning and coding tasks. Terra is the middle tier, competitive with GPT 5.5 at half the cost. Luna is the cheapest and fastest, priced at $1 per million input tokens and $6 per million output tokens. The naming system is new, but the tier structure mirrors Anthropic's Opus-Sonnet-Haiku lineup. OpenAI hinted that this naming convention will stick, so future updates may not jump to "GPT 6" but instead refine within the Sol-Terra-Luna framework.

Continue reading the full summary in the app — free to try.

Read Full Summary →

Free • No credit card required

Listen to Audio Summary Open in App

What you'll learn

  • 1 (00:00) **Introduction: The Unreleased GPT-5.6** - Open AI announced GPT-5.6 but blocked normal ChatGPT access, giving only a few trusted companies access due to federal government requests.
  • 2 (04:28) **The New Three-Tier Model Family: Sol, Terra, Luna** - GPT-5.6 is split into three durable tiers: Sol (flagship), Terra (balanced), and Luna (fast/cheap).
  • 3 (06:26) **Rollout and Pricing Details** - The model is only available via API and Codex to about 20 trusted organizations, not inside ChatGPT.
  • 4 (10:18) **OpenAI's Official Statement and Government Engagement** - OpenAI previewed the model to the U.S. government ahead of launch and agreed to a limited preview at their request.
  • 5 (13:10) **Initial Benchmarks and Performance** - GPT-5.6 Sol beats Mythos V on key benchmarks like Terminal Bench 2.1 (91.9 vs. 88), but matches Mythos Preview on Exploit Bench at a third of the cost.
  • 6 (16:34) **Shopify CTO's Real-World Testing** - Shopify CTO Mikhail Parakhin tested both Fable 5 and GPT-5.6, finding GPT-5.6 clearly better than Opus 4.8 but worse than Fable 5 on coding.
  • 7 (18:26) **New Max and Ultra Modes; Safety Concerns** - Max mode lets Sol reason longer; Ultra mode runs subagents in parallel, similar to Claude Code.

+ Full timestamped outline available in the app

Show Notes

Good news: OpenAI's GPT-5.6 has been released! 🥳
Bad news: 99.9% of users can't access it. 😥

While OpenAI has had to change their release approach due to increased government involvement in AI oversight, the majority of AI leaders are stuck in a new bind: 

Sitting and waiting for AI access. 

Since Anthropic's Fable fights with the U.S. government, the Trump admin has increased its regulatory oversight. 

But don't sit and wait. Because when availability opens to all, your org will need to be ready. We'll help you understand the lay of the new land. 


Newsletter: Sign up for our free daily newsletter
More on this Episode: Episode Page
Today's Episode on LinkedIn: Thoughts on this? Join the convo on LinkedIn and connect with other AI leaders.

Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineup
Website: YourEverydayAI.com
Email The Show: [email protected]
Connect with Jordan on LinkedIn

Topics Covered in This Episode:

  1. OpenAI GPT-5.6 Announcement & Delay
  2. Federal Government’s AI Model Review Process
  3. GPT-5.6 Model Family: Sol, Terra, Luna Explained
  4. New Durable Product Tiers in OpenAI Models
  5. Breakdown of GPT-5.6 Sol, Terra, Luna Pricing
  6. GPT-5.6 Benchmarks: Coding and Agentic Tasks
  7. Safety, Cybersecurity, and Bio-Risk Capabilities
  8. Anthropic Shutdown Compared to OpenAI Launch
  9. US-China AI Race and Model Distillation Issues
  10. Enterprise AI Supply Chain & Access Risks
  11. Operator Playbook: Model Fallbacks and Routing




Timestamps:

00:00 Delayed GPT-5.6 release concerns

04:37 Explaining GPT 5.6 Model Tiers

08:50 Luna usage by developers

11:08 Previewing new model with US government

12:53 OpenAI's government collaboration concerns

15:55 Cutting Costs on AI Tokens

19:02 OpenAI model safety concerns

24:13 Anthropic's export order issue

27:46 Accessing the Start Here series

30:11 US enterprises and Chinese open source models

34:10 Managing AI model access risks

36:41 Navigating AI model restrictions





Everyday AI Podcast – An AI and ChatGPT Podcast

More from this podcast

Everyday AI Podcast – An AI and ChatGPT Podcast →