Apple’s new machine learning framework thinks the old Mac Pro is a space heater

“During the latest edition of its annual WWDC event, Apple made important strides to show developers and creators that it is finally getting serious about artificial intelligence,” MIX reports for TNW.

“The company announced its all-new Core ML framework specifically designed to enable developers to build smarter apps by embedding them with on-device machine learning capabilities,” MIX reports. “But it seems the new system still has some learning to do.”

“Toying around with the Core ML beta, developer Paul Haddad took to Twitter to showcase how well the framework handles computer vision tasks,” MIX reports. “Using the new built-in screen recording tool, Haddad tested the capacity of Core ML to identify and caption objects in real-time.”

“Despite little inconsistencies, Haddad expressed enthusiasm about the framework’s potential, noting that users need to point their cameras ‘at the right angle’ for optimal results,” MIX reports. “Given that apps relying on machine learning algorithms to caption objects in the real world misidentify their targets all the time, it is hardly surprising Apple’s framework is getting stuff wrong here and there…”

Read more in the full article here.

MacDailyNews Take: Beta.

At least it didn’t call it a cheese grater.

And, depending on the CPU and GPU installed, the Core ML beta is actually correct.


  1. Mac Pro cheese graters were definitely floor space heaters. And in a small office it could get quite hot and you’d have to crank up the AC to compensate. Hopefully the next 2018 Mac Pro’s might be a little less thermally volatile.

    1. Ideally they will find a balance between the cylindrical Mac Pro’s lower heat generation and the older Mac tower’s raw power. Pro users who need all the processing power available, won’t mind a little extra heat and/or will have the AC on hand to compensate.

  2. If it’s called ‘AI’, you can count on it being clunky. We’re still in the dark age of computing, most especially with AI.

    BUT! I like the current progress. It’s starting to get interesting, rather than just rehashing and improving on very old ‘AI’ technology.

Reader Feedback

This site uses Akismet to reduce spam. Learn how your comment data is processed.