Just Months After Jeopardy!, Watson Wows Doctors With Medical Knowledge

[Snow Leopard]

Computers are becoming increasingly capable, being superior than humans at playing chess, Jepoardy, driving your Prius, flying planes and now diagnosing and treating disease. Or at least they will in the foreseeable future.

http://singularityhub.com/2011/06/0...y-watson-wows-doctors-with-medical-knowledge/

I believe IBM also has plans to make lawyers obsolete.

 

[biophile]

Sounds like a poll topic. "Would you prefer to see Watson over the average medical doctor for your ME/CFS?" Possible issue before tempting to answer "hell yes!": GIGO even worse for Watson, could be fed with CDC data and all 3 paragraphs of text from general medical textbooks, not to mention countless reams of psychobabble! Would be funny though if Watson acquired the ability to judge papers on their scientific merit and start rejecting the flawed ones. Johnny 5 is alive and needs more input!

 

[alex3619]

Hi, I was in the AI field in the early 90s. They already had robotic surgeons that could do sensitive surgery better than humans - but the legal implications terrified the developers. Who is liable if a robot makes a mistake? Bye, Alex

 

[Desdinova]

H E double hockey sticks NO! I could hear the medical community now "But the computer says...". As Biophile said it would be a major case of GIGO. Since it would more then likely focused on major consensus based largely on the volume of evidence.

And since the bulk of reports and studies are garbage ones flooded out by the psychiatrist, government health agencies and government back researchers the conclusion would be inevitable. A computer still can't take into consideration the human factor where medicine, medical and health issues are involved. It's not going to take into account political motivations, economic motivations nor all the patients motivations and problems.

 

[Snow Leopard]

A computer still can't take into consideration the human factor where medicine, medical and health issues are involved.

Right now that is true, but it's not true in principle.

As for legal issues, given that risk in many fields is already been redirected away from the individual towards corporations, I'm sure they'll figure it out. That's probably not what you wanted to hear though, so the point is that the robot/computer will be much less likely to make a mistake than a human and all decisions will be signed off by a nurse at the very least.