Welcome back, AI’ers!

Here’s What’s On The Menu For Today

• OpenAI's GPT-5 model family leaked

• ChatGPT's $1 deal for the US government

• Google’s new AI coding agent, Jules

• Microsoft’s self-reflecting AI framework

AI ON MY MIND

A MESSAGE FROM OUR PARTNER

The Enterprise Guide to Secure Voice AI Rollouts

Deploying Voice AI in a regulated industry? This guide shows how security isn’t just a requirement—it’s your rollout strategy.

Learn how HIPAA and GDPR compliance can accelerate adoption, reduce risk, and scale across 100+ locations.

From encryption and audit logs to procurement readiness, this guide outlines what enterprise IT, ops, and CX teams need to launch AI voice agents with confidence.

Download The Free Guide

TODAY’S TOP STORIES

GPT-5 Is Here

The Recap: A page on GitHub Models appears to have prematurely announced OpenAI's next flagship model, GPT-5. The leak details major capability upgrades and a full family of new model variants ahead of any official release.

Unpacked:

• The new model is described as a major leap forward, with enhanced reasoning, higher-quality code generation, and powerful agentic capabilities.

• The leak reveals four distinct variants designed for different use cases: a primary model for logic, a 'mini' for cost-effectiveness, a 'nano' for speed, and a 'chat' version for enterprise conversations.

• Developers can apparently test the new models directly in the GitHub Models playground, comparing them against offerings from Meta, Cohere, and Microsoft.

Shortly after OpenAI made the announcement on X.

Bottom line: This accidental reveal signals a major escalation in the AI development race, immediately raising the bar for competitors. The specialized model variants show OpenAI is aiming to embed its technology across the entire professional landscape, from high-speed applications to complex enterprise workflows.

ChatGPT For $1

The Recap:  In an aggressive strategic push, OpenAI is offering ChatGPT Enterprise to all U.S. federal agencies for just $1 per agency for the next year. This move aims to undercut rivals and accelerate long-term government adoption.

Unpacked:

• The deal arrives just a day after the General Services Administration (GSA) added OpenAI, Google, and Anthropic to its list of approved AI vendors for federal contracts.

• The offer includes unlimited use of advanced models for an additional 60-day period and access to a government user community with tailored training resources.

• To address data privacy, the GSA confirmed it is taking a “cautious, security-first approach” to ensure sensitive government information remains protected.

Bottom line: OpenAI is making a huge play to become the default AI for the U.S. government, forcing competitors to either match the price or risk being left out. This move positions the company to deeply integrate into federal operations, potentially shaping public sector AI use for years to come.

MY FAVORITE AI TOOLS

Relevance is my favorite AI tool right now (even more so than N8N).

Here’s why:

It’s an AI agent building tool that doesn’t require you to build out the logic using nodes. Instead, you use natural language to build out these agents. But the best thing imo is that you can build AI Agent teams.

Like agents talking to each other completing a goal until its done. It’s nuts. No other tool on the market is doing this. You have to check it out.

(Try it out →Link)

👾 AI Tools We Use Everyday

➡️ N8N for AI Agents - Literally use this everyday to write this newsletter (link)

➡️ Axiom AI - AI Browser automation to scrape anything on the web (link)

➡️ GummySearch - AI Reddit research (link)

Tools are sourced from our database and are not sponsored or affiliated.

MORE STORIES

Google's New Coding Agent

The Recap: Google has moved its coding agent Jules out of beta, making the AI assistant generally available to help both developers and enthusiasts write, test, and improve code.

Unpacked:

• Jules sets itself apart by running asynchronously in a virtual machine, which lets you assign complex tasks, close your laptop, and return later to the completed work.

• The official release includes new features like multimodal support for visual outputs, the ability to reuse past setups, and direct GitHub Issues integration.

• A free version allows for 15 individual tasks per day, while unlimited use requires a Google AI Pro or Ultra subscription.

Bottom line: Jules represents Google's significant push into agentic AI, moving beyond simple chat interfaces. This shift points to a future where AI assistants proactively manage and complete complex workflows, acting as autonomous teammates.

Microsoft's Self-Reflecting AI

The Recap: Microsoft has introduced CLIO, a new research framework that enables large language models to adapt and refine their own thought processes in real-time. This self-correction capability shows significant performance gains in complex scientific problem-solving.

Unpacked:

• The new framework, named Cognitive Loop via In-situ Optimization (CLIO), allows models to adjust their internal reasoning as they work on a task.

• It works by creating self-reflection and feedback loops, which lets an LLM evaluate and improve its approach without external training.

• In early tests, CLIO boosted GPT-4.1's performance on challenging biomedical questions, demonstrating its potential for high-stakes applications.

Bottom line: This framework pushes AI from simply executing tasks to actively improving its own methods during execution. Such self-adapting models could soon tackle specialized scientific and technical challenges with greater accuracy and autonomy.

OTHER NEWS YOU MAY LIKE

Universal Pictures warned it will sue tech companies that use its films for AI training, adding explicit legal threats to the credits of its new movies.

Google committed $1 billion over three years to support AI training in higher education, providing free access to its AI Pro plan for college students in the U.S. and four other countries.

The Browser Company launched a $20/month Pro plan for its AI-native browser, Dia, introducing its first subscription model to monetize the browser's advanced AI features.

xAI stunned competitors in Google's AI chess tournament, with its Grok 4 model demonstrating impressive strategic play despite not being specifically trained on the game.

SPONSOR US

Get your product in front of over 46k+ tech & AI enthusiasts

Our newsletter is read by thousands of tech professionals, investors, engineers, managers, and business owners worldwide. Get in touch to learn more.

Need Help?

📩 Need help with an AI idea / project?

Have something you’re trying to build, automate, or scale using AI?

Let my team help you figure it out.

👉 Submit your idea or request here

We’ll take a look and get back to you with suggestions, resources, or a potential implementation plan — no strings attached.

🚀 Want to Make Money from AI?

We’re quietly launching a private beta program that trains a small group to become AI Implementers — People who help other founders install AI systems and get paid for it.

If you're interested in joining this early group…

👉 Reply with the word “implementer”

AI ON SOCIAL

 AI ART

RANDOM PROMPT OF THE DAY

Want more prompts?

We have a database of over 8,000 of the best prompts we’ve collected, which you can access by sharing Best of AI using the link below.

THAT’S IT FOR TODAY

Until next time, stay safe.

-Best of AI Team

Remember to click to unsubscribe if you don’t want us to land in your inbox anymore.