3 ms·

I'm genuinely curious - can you give some concrete examples of what you consider to be such BS ML systems? It would be fun in a TheDailyWTF way :)

baron_harkonnen 7y ago
I would really enjoy that but am still a little too close to these systems to give the delightful and horrifying accounts that I would like to.
buzzier 7y ago
AI Circus, Mid 2019 Update https://blog.piekniewski.info/2019/05/30/ai-circus-mid-2019-update/ https://blog.piekniewski.info/2019/05/30/ai-circus-mid-2019-... Stories of AI Failure and How to Avoid Similar AI Fails in 2019 https://www.lexalytics.com/lexablog/stories-ai-failure-avoid-ai-fails-2019 https://www.lexalytics.com/lexablog/stories-ai-failure-avoid... How IBM Watson Overpromised and Underdelivered on AI Health Care https://spectrum.ieee.org/biomedical/diagnostics/how-ibm-watson-overpromised-and-underdelivered-on-ai-health-care https://spectrum.ieee.org/biomedical/diagnostics/how-ibm-wat... Automakers Are Rethinking the Timetable for Fully Autonomous Cars https://www.designnews.com/electronics-test/automakers-are-rethinking-timetable-fully-autonomous-cars/93993798360804 https://www.designnews.com/electronics-test/automakers-are-r...
- Fr0styMatt88 7y ago
  Just what I need for my work commute, thanks! :) As an aside, my worry with AI/ML is that we’re creating these giant, un-debuggable systems that can’t be fixed when specific problems arise. Every time I see complaints about YouTube/Facebook/<some other company> giving nonsense recommendations to people I think “How would they even fix that? Do they even know how their recommendation systems work?”. One of the great things about software is that when something is broken, you can go right in there and fix it. With ML, I can imagine getting to a point where our only option is to take the black-box we’ve created to therapy and hope for the best.... :)
staticassertion 7y ago
Look at security startups.

baron_harkonnen 7y ago

I would really enjoy that but am still a little too close to these systems to give the delightful and horrifying accounts that I would like to.

buzzier 7y ago

AI Circus, Mid 2019 Update https://blog.piekniewski.info/2019/05/30/ai-circus-mid-2019-update/ https://blog.piekniewski.info/2019/05/30/ai-circus-mid-2019-... Stories of AI Failure and How to Avoid Similar AI Fails in 2019 https://www.lexalytics.com/lexablog/stories-ai-failure-avoid-ai-fails-2019 https://www.lexalytics.com/lexablog/stories-ai-failure-avoid... How IBM Watson Overpromised and Underdelivered on AI Health Care https://spectrum.ieee.org/biomedical/diagnostics/how-ibm-watson-overpromised-and-underdelivered-on-ai-health-care https://spectrum.ieee.org/biomedical/diagnostics/how-ibm-wat... Automakers Are Rethinking the Timetable for Fully Autonomous Cars https://www.designnews.com/electronics-test/automakers-are-rethinking-timetable-fully-autonomous-cars/93993798360804 https://www.designnews.com/electronics-test/automakers-are-r...

Fr0styMatt88 7y ago
Just what I need for my work commute, thanks! :) As an aside, my worry with AI/ML is that we’re creating these giant, un-debuggable systems that can’t be fixed when specific problems arise. Every time I see complaints about YouTube/Facebook/<some other company> giving nonsense recommendations to people I think “How would they even fix that? Do they even know how their recommendation systems work?”. One of the great things about software is that when something is broken, you can go right in there and fix it. With ML, I can imagine getting to a point where our only option is to take the black-box we’ve created to therapy and hope for the best.... :)

staticassertion 7y ago

Look at security startups.