Why AI Hallucinates - And How IBM Is Solving It

AI's dirty little secret just got a corporate fixer.

IBM cracks the code on machine imagination gone wild

Forget everything you've heard about AI perfection - these systems spin fiction faster than a crypto bro pitching the next shitcoin. IBM's new architecture tackles the hallucination problem head-on, implementing verification layers that actually make AI accountable for its outputs.

The solution doesn't just patch holes - it rebuilds the foundation. Real-time fact-checking modules cross-reference claims against verified databases before responses ever reach users. No more making up financial projections or inventing regulatory frameworks.

Suddenly, AI might actually be trustworthy enough for enterprise applications beyond generating marketing fluff. IBM's approach could finally separate artificial intelligence from artificial creativity - giving businesses tools that analyze rather than fantasize.

Turns out the real innovation wasn't teaching machines to think, but teaching them to tell the truth.

Elevate Your Investing Strategy:

• Take advantage of TipRanks Premium at 50% off! Unlock powerful investing tools, advanced data, and expert analyst insights to help you invest with confidence.

As a result, this encourages models to give answers instead of admitting they don’t know something, which experts say creates the wrong incentives. More specifically, Ayhan Sebin from tech giant IBM (IBM) compares it to human performance metrics, saying that if the system rewards guessing, then models will guess. Separately, Kate Soule, another IBM expert, says that this is a “calibration problem.” Right now, AI systems are pushed to always give an answer, which leads to more hallucinations. But going too far in the other direction, where models say “I don’t know” too often, also makes them less useful.

That’s why researchers believe that better ways to score AI models are needed, so they learn when to answer and when to stay silent, depending on the situation. Therefore, IBM is exploring solutions, like a new system called Larimar, which gives models editable memory so they can learn and revise facts in real time. Instead of retraining the whole model, Larimar lets AI update specific information. While hallucinations won’t disappear right away, tools like Larimar and better testing methods could help keep them under control.

Is IBM a Buy, Sell, or Hold?

Turning to Wall Street, analysts have a Moderate Buy consensus rating on IBM stock based on seven Buys, seven Holds, and one Sell assigned in the past three months, as indicated by the graphic below. Furthermore, the average IBM price target of $294.86 per share implies 13.8% upside potential.