Digital Digest #20 - Drowning in disinformation; content moderation; Threads and news; gen AI hallucinations

Hi everyone,

This is a weird time to be writing a digital newsletter. The types of topics we often cover here feel so insignificant with everything going on right now.

I was horrified last week by the brutal attacks against Israel, and heartbroken by the humanitarian crisis that is now unfolding. My thoughts and sympathies are with everyone who has been affected.

But there are some practical implications of all of this for digital communicators. I've had innumerable conversations with brands over the last week relating to whether to speak publicly and what to say if/when they do; whether to pause content; whether to proceed with marketing or influencer campaigns; how to work with influencers who have their own perspectives; how to deal with misinformation about their organizations; the list goes on.

So, with a heavy heart, I present this week's Digital Digest. Most of this week's newsletter is informed or influenced by the past week's events.

Let's get into it.

1. Drowning in disinformation

One major theme over the last week was the prevalence of mis- and disinformation relating to the evolving events in Israel & Gaza. Manipulated images; images from video games; and content from past conflicts have all been shared online under false pretenses in recent days. (Wired, Axios, Insider Intelligence)

• Why it matters: The threat of mis- and disinformation around this situation has not been limited to nation states. Several companies have found themselves facing misinformation and subsequent backlash related to their positions on the conflict. This is likely to continue, and companies who have been outspoken on the issue, who have faced advocacy or activism related to this region or who have faced prior misinformation issues would be well advised to step up their monitoring for these types of threats.
• Also: TikTok accounts are spreading falsehoods with help from A.I.-generated voices, according to the New York Times (NYT)

2. Platforms wrestle with content moderation

Digital platforms have been used as weapons in the information war around the ongoing conflict, and some have struggled to keep up with the onslaught of graphic content - leading to the EU warning multiple platforms and encouraging them to "step up" their efforts. Both Meta and TikTok in recent days have published outlines of how they've been working to protect users. (Politico, BBC, TikTok, Meta)

• Why it matters: In this volatile landscape - and often out of a desire to be respectful of appropriateness in the current context - many brands paused publishing of digital content and campaigns over the last week. As the situation continues to evolve, decisions over whether and how to resume posting will remain challenging and subjective in the near term.

3. Threads won't amplify news

While X (Twitter) fell short in the first major test of its new regime's approach to handling contentious news events (not helped by its utterly bizarre removal of news headlines from shared links), Threads - its rising competitor - remains steadfastly standoff-ish in its approach to news despite the opportunity presented to displace X as the place to go for real-time updates. (TechCrunch, Social Media Today)

• Why it matters: While Instagram head Adam Mosseri insists that the platform is "not anti-news", he stated that the platform will not amplify news on the platform, either. That probably won't stop journalists who are flocking to build their presence on Threads, though.
• Meanwhile: X rolled out a new ad format that doesn't disclose the advertiser or even that it is an ad, and can't be reported or blocked. (Mashable)

4 . Generative AI tools falsely claim ceasefire

"Hallucinations" have been one of the major problems plaguing LLM-driven chatbots since they first came on the market last year. This was given a spotlight this week, when two such tools falsely claimed there was a ceasefire between the two sides of the conflict. (Bloomberg)

• Why it matters: Companies are continuing to grapple with falsehoods shared by chatbots, which represent a major barrier to adoption of these tools within organizations.

Also in generative AI:

• Google promises to take the legal heat in users’ AI copyright lawsuits (Google, The Verge)
• Adobe is experimenting with AI-powered generative fill in videos, along with new models that can create high-quality images, editable vector graphics, and customizable design templates. (The Verge)

That's all for this week. Hope you all stay safe and well.

Dave