Ex-OpenAI CTO solves AI’s biggest bug

PLUS: OpenAI's $300B cloud deal and Replit's autonomous agent upgrade

Together with

A major breakthrough in AI reliability has just emerged from ex-OpenAI CTO Mira Murati's startup. They've released an open-source solution to fix the inconsistency problem that causes AI models to give different answers to the same question.

This development could transform how businesses and researchers rely on AI systems for critical tasks. With consistent, reproducible results now possible, are we entering a new era of truly dependable artificial intelligence?

Today in AI:
  • Ex-OpenAI CTO solves AI's consistency bug

  • OpenAI's historic $300B Oracle cloud deal

  • Replit's coding agent gets a major upgrade

Protect your checkout from coupon plug-ins. Boost your margin today.

KeepCart: Coupon Protection partners with DTC brands like Quince, Blueland, Vessi and more to protect your checkout from plug-ins like Honey, CapitalOne, RetailMeNot, and more to boost your DTC margins

Overpaid commissions to affiliates and influencers add up fast – Get rid of the headache and revenue losses with KeepCart.

After months of using KeepCart, Mando says “It has paid for itself multiple times over.”

Now it’s your turn to see how much more profit you can keep.

What’s new? Ex-OpenAI CTO Mira Murati’s new startup, Thinking Machines Lab, has published research and code to solve nondeterminism, a persistent bug that causes AI models to produce different outputs for the same prompt.

What matters?

  • The problem, called nondeterminism, means models give inconsistent answers even at their most predictable settings, which undermines reliability for critical business applications and scientific research.

  • The team discovered the issue isn't random math errors but "batch invariance," where a model's output changes based on how many other user requests are processed alongside it—like your coffee order tasting different during a rush.

  • The startup has already released its solution as open-source code, allowing developers everywhere to implement "batch-invariant" operations and build more predictable AI models immediately.

Why it matters?

This breakthrough addresses a fundamental flaw in AI, paving the way for models that are more trustworthy and reproducible for enterprise-scale tasks. By releasing the solution openly, Thinking Machines Lab empowers the entire industry to advance toward more dependable and consistent AI systems.

What’s new? Oracle and OpenAI have signed a historic $300 billion, five-year cloud contract, securing a massive amount of AI computing capacity to power future models.

What matters?

  • The deal, which commits OpenAI to spending roughly $60B annually starting in 2027, is part of a larger $455B in AI infrastructure contracts Oracle revealed.

  • The news caused Oracle’s stock to surge 40%, which briefly made co-founder Larry Ellison the world's richest person.

  • This agreement provides the massive power needed for ambitious OpenAI initiatives like Project Stargate, cementing Oracle’s position as a critical infrastructure provider in the AI gold rush.

Why it matters?

This massive investment underscores that raw computational power is now the most critical and contested resource in the race for next-generation AI. It signals that the biggest financial wins may come not just from building models, but from supplying the foundational infrastructure that powers them.

How 433 Investors Unlocked 400X Return Potential

Institutional investors back startups to unlock outsized returns. Regular investors have to wait. But not anymore. Thanks to regulatory updates, some companies are doing things differently.

Take Revolut. In 2016, 433 regular people invested an average of $2,730. Today? They got a 400X buyout offer from the company, as Revolut’s valuation increased 89,900% in the same timeframe.

Founded by a former Zillow exec, Pacaso’s co-ownership tech reshapes the $1.3T vacation home market. They’ve earned $110M+ in gross profit to date, including 41% YoY growth in 2024 alone. They even reserved the Nasdaq ticker PCSO.

The same institutional investors behind Uber, Venmo, and eBay backed Pacaso. And you can join them. But not for long. Pacaso’s investment opportunity ends September 18.

Paid advertisement for Pacaso’s Regulation A offering. Read the offering circular at invest.pacaso.com. Reserving a ticker symbol is not a guarantee that the company will go public. Listing on the NASDAQ is subject to approvals.

What’s new? AI coding platform Replit announced a $250M funding round, boosting its valuation to $3B, and unveiled Agent 3, its most autonomous agent yet for building and testing applications.

What matters?

  • Agent 3 can operate for up to 200 minutes without human intervention, a 10x increase in autonomous runtime compared to its predecessor.

  • Beyond coding, the agent can create workflow automations and auto-test applications by simulating user interactions directly in a browser.

  • The company’s reported revenue has surged from $3M to $150M in the past year, signaling strong market confidence in the agentic coding sector.

Why it matters?

Replit's progress highlights the accelerating capability of AI agents to handle complex, long-duration software development tasks from start to finish. This shift points toward a future where developers act more as project managers for AI agents, significantly speeding up the creation and deployment of new applications.

Everything else in AI

Anthropic enabled its Claude model to generate and create files directly within conversations, including PowerPoint presentations, Excel spreadsheets, and PDFs.

Publishers launched the Real Simple Licensing (RSL) protocol, a new standard backed by Reddit, Yahoo, and Medium to control and monetize how AI companies scrape and use their content.

Meta paid $140M to license AI image generation technology from startup Black Forest Labs, marking a strategic shift toward integrating external tech rather than relying solely on in-house development.

Researchers unveiled SpikingBrain 1.0, a brain-inspired AI model from China that is reportedly 25-100 times faster than transformers while using only 2% of the typical training data.

The Gold standard for AI news

AI will eliminate 300 million jobs in the next 5 years.

Yours doesn't have to be one of them.

Here's how to future-proof your career:

  • Join the Superhuman AI newsletter - read by 1M+ professionals

  • Learn AI skills in 3 mins a day

  • Become the AI expert on your team

Stop Struggling with AI. This all-in-one bundle gives you 1,000+ battle-tested MidJourney, ChatGPT, Claude, Gemini & Grok prompts. You can create stunning visuals, logos, automate tasks and much more! Plus: Lifetime of access.

Why This Bundle Changes Everything:

 Instant Expertise — Skip $1000s in consultant fees with ready-to-use systems
 Massive Time Recovery — Automate repetitive tasks that drain your day
 Proven Formulas — True results without guesswork
 Future-Proofed — Stay ahead with lifetime updates at no extra cost

Special Offer: Use VISTA to get Extra 10% off at the checkout!

Essential AI Guides - Reading List:

Your opinion matters!

Work with us

Reach 100k+ engaged Tech Professionals, Engineers, Managers and decision makers. Join brands like MorningBrew, HubSpot, Prezi, Nike, Ahref, Roku, 1440, Superhuman, and others in showcasing your product to our audience. Get in touch now →