With the somewhat deceptive “AI washing” in software marketing lately, it’s important to ask: How far have we really come toward true artificial intelligence?

“There’s a great deal of difficulty in trying to code the human brain,” said Janet George (pictured), fellow and chief data officer at Western Digital Corp. Products claiming to match human intelligence likely underestimate the huge complexity and power of the human brain, she said during this year’s When IoT Met AI: The Intelligence of Things conference in San Jose, California.

Machine learning peppers big data conversations lately, but it’s a cut beneath neural networks in terms of true AI, George told Jeff Frick (@JeffFrick), host of theCUBE, SiliconANGLE Media’s mobile livestreaming studio. (* Disclosure below.)

Both ML and neural network algorithms iteratively learn from data, but there are key differences.

“We’re really dealing with mathematical approximation. We’re not dealing with preciseness. We’re not dealing with exactness,” George said of ML. An ML model must be pre-tuned in multiple ways; it could be over-fitted, under-fitted or biased. Features of the data must be extracted in order to make predictions, and this can take an enormous amount of time, according to George.

Neural networks do not require much garbage-in before they can generate insight from data. “You can throw all the raw data at it. It’s in fact data agnostic,” she said.

Anomalous driving

One caveat: Neural networks do require some help with the “failed” data that does not produce an accurate prediction. “If there are no labels on the failed data, it’s too difficult for the neural networks to figure out what the failure is,” George said.

Once a neural network learns an anomaly, it can integrate it with the rest of its makeup. Autonomous vehicles with neural networks, for example, can potentially learn to avoid all types of crashes this way, she stated.

Watch the complete video interview below, and be sure to check out more of SiliconANGLE’s and theCUBE’s coverage of When IoT Met AI: The Intelligence of Things. (* Disclosure: TheCUBE is a paid media partner for When IoT Met AI. Neither Western Digital Corp., the event sponsor, nor other sponsors have editorial influence on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE