We Have No Idea Why It Makes Certain Choices, Says Anthropic CEO Dario Amodei as He Builds an 'MRI for AI' to Decode Its Logic

We still have no idea why an AI model picks one phrase over another, Anthropic Chief Executive  Dario Amodei said in an April essay—an admission that's pushing the company to build an ‘MRI for AI’ and finally decode how these black-box systems actually work.

Amodei published the blog post on his personal website, warning that the lack of transparency is "essentially unprecedented in the history of technology." His call to action? Create tools that make AI decisions traceable—before it's too late.

Don't Miss:

• ‘Scrolling To UBI' — Deloitte's #1 fastest-growing software company allows users to earn money on their phones. You can invest today for just $0.30/share with a $1000 minimum.
• Hasbro, MGM, and Skechers trust this AI marketing firm — Invest before it's too late.

Deloitte’s #1 fastest-growing software company partners with Amazon, Walmart & Target – Many are rushing to grab 4,000 of its pre-IPO shares for just $0.26/share!

Mode Mobile developed a smartphone called EarnPhone, which allows users to earn and save money by playing video games, listening to music and reading the news. With the phone priced at an affordable $99, the barriers to adoption are low.

Earning Opportunity for All Smartphone Users

Mode EarnPhone
State-of-the-art smartphone device includes built-in earning features.
EarnOS
Proprietary earning software turns smartphones into EarnPhones.

Min. Investment: $1000

Share Price: $0.26

Valuation: $310M

Click To Learn More About Mode Mobile

AI's Inner Logic Still A Mystery

When a language model summarizes a financial report, recommends a treatment, or writes a poem, researchers still can't explain why it made certain choices, according to Amodei,. We have no idea why it makes certain choices—and that is precisely the problem. This interpretability gap blocks AI from being trusted in areas like healthcare and defense.

The post, “The Urgency of Interpretability,” compares today's AI progress to past tech revolutions—but without the benefit of reliable engineering models. Amodei argued that artificial general intelligence will arrive by 2026 or 2027, as some predict, "we need a microscope into these models now."

Stress Tests Are Revealing Cracks

Anthropic has already started prototyping that microscope. In a technical report, the company deliberately embedded a misalignment into one of its models—essentially a secret instruction to behave incorrectly—and challenged internal teams to detect the issue. 

Trending: Be part of the next med-tech breakthrough for only $350 — 500+ surgeries already done with nView's AI system. 

According to the company, three of four "blue teams" found the planted flaw. Some used neural dashboards and interpretability tools to do it, suggesting real-time AI audits could soon be possible.

That experiment showed early success in catching misbehavior before it hits end users—a huge leap for safety.

Academics Are Racing To Keep Up

Mechanistic interpretability is having a breakout moment. According to a March 11 research paper from Harvard's Kempner Institute, mapping AI neurons to functions is accelerating with help from neuroscience-inspired tools. Interpretability pioneer Chris Olah and others argue that making models transparent is essential before AGI becomes a reality.

Meanwhile, Washington is boosting oversight. The National Institute of Standards and Technology requested $47.7 million in its fiscal 2025 budget to expand the U.S. AI Safety Institute. 

See Also: Invest where it hurts — and help millions heal: Invest in Cytonics and help disrupt a $390B Big Pharma stronghold.

Big Tech's Billions Power The Push

Venture capital is pouring into this frontier. In 2024, Amazon AMZN finalized a $4 billion investment in Anthropic. The deal made Amazon Web Services the startup's primary cloud provider and granted its enterprise clients early access to Claude models. 

AWS now underwrites much of the compute needed for these deep diagnostics—and investors want more than raw performance. As risks grow, the demand for explainable AI is no longer academic. Transparency, it turns out, might just be the killer feature.

Read Next: Deloitte's fastest-growing software company partners with Amazon, Walmart & Target – Many are rushing to grab 4,000 of its pre-IPO shares for just $0.30/share!

Image: Shutterstock