4 ms·
This is wholly unsurprising. IBM's big play was to integrate data science methods into the workflow. But they approached it from a "we will replace your labor c
by tomrod 5y ago
This is wholly unsurprising. IBM's big play was to integrate data science methods into the workflow. But they approached it from a "we will replace your labor costs" versus "we will augment your labor costs." Besides their AI models being fairly poor in quality, technology doesn't replace people very well where extrapolation is needed. So the quality of service Watson brought was significantly lower than what these businesses offered prior to adoption. So keeping Watson became an exercise in how well the business understands sunk costs and switching costs.
- lvl100 5y agoThis so much. They basically let Snowflake (among others) eat their lunch. Back in 2015, IBM actually had a good chance but leadership just dropped the ball.