2019.02.22 Last Week in Digital Media

Yes, it’s another Last Week in Digital Media “The Late Edition” as I have missed my usual late Sunday afternoon publishing deadline. I couldn’t miss this week given all the news around Brand Safety, so starting with that news first.

Reader Advisory: The news related to this most recent YouTube incident includes references to predatory behavior and disturbing content. All links are to relevant editorial or news coverage only, but reader discretion is advised.

  • On Monday (2/18) a reddit post by Matt Watson identified predatory behavior on YouTube targeting preteens. Watson also made a YouTube video summarizing his findings, which went from 842,000 views on Monday to just shy of 3MM by the time of writing this Sunday (2/24).
  • The behavior Watson identified is criminal and unacceptable. In the case of YouTube, the problem was exacerbated because ads were running in front of the content. The response from the advertising community was swift – with numerous advertisers suspending their YouTube advertising.
  • While confidentiality means I can’t share specifics of actions taken to protect advertisers, I can say Google responded rapidly to the concerns raised. Even if it is more-than-frustrating that Google didn’t catch and prevent this themselves.
  • Publicly, Google confirmed over 400 channels were deleted, as too were 10’s of millions of comments, as well as suspending monetization (Google keeps updating the content on that link, so it’s worth bookmarking and checking back often). The latter two (2) steps were also taken on videos that could be at risk of being targeted by predators. Some innocent channels and content were deleted as Google stated it was acting “in an abundance of caution”.
  • It is clear that Google understood the gravity of the situation, not just from an advertising perspective but the full legal and ethical implications. This includes reporting illegal behavior to NCMEC so that proper action can be taken by relevant law enforcement. In Google’s words “No form of content that endangers minors is acceptable to us.”
  • Just a month prior, AT&T announced they were returning to YouTube, but in light of the recent incident a spokesperson for AT&T told the BBC “Until Google [which owns YouTube] can protect our brand from offensive content of any kind, we are removing all advertising from YouTube“.
  • Unrelated to the above issue but coincidentally timed during the same week (2/19), Google did update the YouTube strike system that could result in channel termination. Now 3 strikes in any 90 day period can result in account termination. The new rules come into effect Feb 25.

Editorial: The past week underscores the need for further improvements in how 3rd Party Verification works on YouTube and on all User Generated Content (UGC) environments. There is also a real need for continued safety improvements by Google themselves. I am deliberate when I call this “safety”, not “brand safety”, as, without question, the type of content and behavior has no place (regardless of advertising). From an agency perspective, there are legitimate questions on why some of the content passed safety checks, how YouTube classified some of the content, and why it allowed it to be monetized. All of this needs to be addressed and you should talk to your relevant Google/YouTube account lead to understand what steps are being taken.

Looking ahead, the abuse of the comment system is disturbing because it doesn’t require sophisticated techniques to make content “unsafe”. YouTube is already struggling to combat “dislike mobs” who mass thumb down videos. Inappropriate comments, dislike mobs, etc. are behaviors that are easy to replicate with bots (see what happened to the FCC and Net Neutrality comments system). I am not an apologist for or excusing YouTube’s problems from last week. I am left wondering what bad actors will take from this event and use it to undermine legitimate content or debate by weaponizing comments and other engagement tools.

  • All of the above wasn’t the only incident that occurred on YouTube during the week. Buzzfeed also highlighted that anti-vaccination videos were being promoted. The Buzzfeed article came off a letter the previous week by Congressman Adam Schiff (PDF link) asking what Google was doing to stop surfacing disinformation. All of which culminated in a YouTube blog post later in the week (somewhat lost in all the other issues) detailing efforts being made to “reduce recommendation of content that could misinform in harmful ways“.
  • If you’re still following all of this, there’s a confluence of issues here – as in both cases, YouTube’s algorithm was recommending similar either inappropriate or misinformed content to users. This demonstrates the issues are not just about content, but also how Google’s systems are geared to effectively amplifying questionable content once a user heads down the wrong path.

On to the other news you may have missed during the week:


Thanks for being patient with the late publishing of the blog this week and I hope you have a great week.


popup close button