back to article Facebook can't root out fake news and hate talk, but – oh look – it has software to catch bugs

Facebook may have to hire people to police the content coursing through its social network, but software looks to be sufficient to hunt down bugs in its mobile app code. On Thursday, the information-harvesting biz revealed but did not release SapFix, a debugging tool that relies on artificial intelligence to suggest fixes for …

  1. onefang

    What can possibly go wrong?

    1. Anonymous Coward
      Anonymous Coward

      What can possibly go wrong?

      The AI starts modifying SapFix and SkyNet is born!

      1. VikiAi
        Terminator

        If SkyNet is born in facebook's systems, then either it will delete itself or kill us all out of compassion.

        1. Anonymous Coward
          Anonymous Coward

          If SkyNet is born from this, its use for "fixing" the Android app was destiny.

        2. onefang
          Thumb Up

          "If SkyNet is born in facebook's systems, then either it will delete itself or kill us all out of compassion."

          I wonder how many likes it'll get from the other AIs on FB when it kills us all?

        3. CrazyOldCatMan Silver badge

          If SkyNet is born in facebook's systems, then either it will delete itself or kill us all out of compassion.

          It won't be from compassion, but more from boredom and malice.

          I have no mouth and yet I must scream.

  2. el kabong

    And their next step will be...

    Put a new bug catcher one level above the first bug catcher to catch bug catcher bugs below.

  3. Anonymous Coward
    Devil

    " so that they can be checked over for"...

    ... any useful behavioural data for targeted advertising.

    Fake news are just a subset of it.

  4. Anonymous Coward
    Anonymous Coward

    I tyhink fake news would be much harder to check for, than figuring out where and when the bug occurred, especially when you have access to the binaries and source. Then it's probably a matter of reverting those lines (assuming they knew the line number) or reverting the whole file. Failing that it seems to use a lookup of sorts to fix it similar to previous fixes.

    Fake news would probably require more advanced artificial intelligence to understand what the post is saying and compare that to what it knows.

    Hate talk should be easier to look into, but would again likely need to understand semantics of language and to differentiate humour (say, racism between friends) and actual hate speech.

    Not that I am condoning FB, I've seen racist anti-white comments on a group which I flagged up,but nothing seemed to have happened. But we shouldn't cloud our judgement because it's FB.

    1. dnicholas

      Fake news is often part truth part wording. Rooting out the context and the intentions of the author requires the sort of intelligence that most PEOPLE on Facebook lack

  5. cNova
    WTF?

    For their Mobile App? Seriously?

    Needing AI to find and "suggest strategies" to patch bugs in your mobile app isn't something one should brag about.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like