3 ms·

Interesting that Watson is cited in the article so heavily, because I strongly believe that the reason Watson failed was because of MBA's and Management Consult

by sgt101 6y ago

Interesting that Watson is cited in the article so heavily, because I strongly believe that the reason Watson failed was because of MBA's and Management Consultants. I could have told them (in fact, as an IBM customer, I did) that going for healthcare was deranged, but they went after the $$$ in complete denial of the difficulties (heavy regulation, safety critical, deeply powerful complex network of stakeholders, super complex badly understood domain). Also they went far to fast, if they had tempered the effort with some patience then they could have got some really decent progress on a simpler domain (like customer service for telco's). But the beast needed feeding.

josefx 6y ago
> in complete denial of the difficulties (heavy regulation, safety critical, deeply powerful complex network of stakeholders, super complex badly understood domain) The last article I read about it cited rather mundane reasons for it ending up unused. Things like not even supporting the data formats used by hospitals that served as cutting edge users.
scotty79 6y ago
I think they didn't get how bizarre is US healthcare. It's a "market" (maybe rather a system) where (nearly) every participant is interested in everything being as expensive as humanely possible. Since automation saves money and Watson is just automation it has no value for US healthcare participants. If they came up with completely new thing that could be priced additionally, it would have been a different talk.
- 256lie 6y ago
  There are healthcare startups around fraud detection, reducing no-shows, telemedicine, drug discovery, and patient triage. Just radiology alone is prime for ML due to existing digital infrastructure and clinical use cases. The trend in FDA cleared AI products is pretty clear over the past decade. https://models.acrdsi.org/ https://models.acrdsi.org/
256lie 6y ago
Watson missed the DL train and IBM should have partnered with a company that had experience in getting medical devices through the FDA (like MSFT are doing with Nuance).
kickout 6y ago
That's what kills me and continues to perplex me. Something like AI for agriculture machines (think maize and soybean fields) are NOT heavily regulated, safety critical, or deeply complex. Its basically 160 acres of open space with reasonable guarantees no humans are around. It would take a trivial amount of time to code in the proper kill switches to deactivate these machines if they detect a human with x feet Sadly this research is only starting to take off (with 1/1000000 of the investment too). 50B-100B industry...
- lumost 6y ago
  Automating heavy machinery is a tricky business. The amortized capex can outstrip the costs of the operator by a significant margin. The cost of downtime due to AI failure may be larger than the savings of not having an operator. If you can replace individual pickers on the other hand...
- sqqqqrly 6y ago
  My grandfather was a farmer in Ohio. He was always very worried that I, as a small child, would get in the field and be hit and killed by his tractor. This may have been because he did accidentally kill a dog once this way.

by sgt101 6y ago

josefx 6y ago

> in complete denial of the difficulties (heavy regulation, safety critical, deeply powerful complex network of stakeholders, super complex badly understood domain) The last article I read about it cited rather mundane reasons for it ending up unused. Things like not even supporting the data formats used by hospitals that served as cutting edge users.

scotty79 6y ago

I think they didn't get how bizarre is US healthcare. It's a "market" (maybe rather a system) where (nearly) every participant is interested in everything being as expensive as humanely possible. Since automation saves money and Watson is just automation it has no value for US healthcare participants. If they came up with completely new thing that could be priced additionally, it would have been a different talk.

256lie 6y ago
There are healthcare startups around fraud detection, reducing no-shows, telemedicine, drug discovery, and patient triage. Just radiology alone is prime for ML due to existing digital infrastructure and clinical use cases. The trend in FDA cleared AI products is pretty clear over the past decade. https://models.acrdsi.org/ https://models.acrdsi.org/

256lie 6y ago

Watson missed the DL train and IBM should have partnered with a company that had experience in getting medical devices through the FDA (like MSFT are doing with Nuance).

kickout 6y ago

That's what kills me and continues to perplex me. Something like AI for agriculture machines (think maize and soybean fields) are NOT heavily regulated, safety critical, or deeply complex. Its basically 160 acres of open space with reasonable guarantees no humans are around. It would take a trivial amount of time to code in the proper kill switches to deactivate these machines if they detect a human with x feet Sadly this research is only starting to take off (with 1/1000000 of the investment too). 50B-100B industry...

lumost 6y ago
Automating heavy machinery is a tricky business. The amortized capex can outstrip the costs of the operator by a significant margin. The cost of downtime due to AI failure may be larger than the savings of not having an operator. If you can replace individual pickers on the other hand...
sqqqqrly 6y ago
My grandfather was a farmer in Ohio. He was always very worried that I, as a small child, would get in the field and be hit and killed by his tractor. This may have been because he did accidentally kill a dog once this way.