Robert F Kennedy Jr is a flake of Cadbury proportions with a famous name. He’s the son of Robert Kennedy, who was assassinated in 1968 when he was running for the Democratic presidential nomination (and therefore also JFK’s nephew). Let’s call him Junior. For years – even pre-Covid-19 – he’s been running a vigorous anti-vaccine campaign and peddling conspiracy theories. In 2021, for example, he was claiming that Dr Anthony Fauci was in cahoots with Bill Gates and the big pharma companies to run a “powerful vaccination cartel” that would prolong the pandemic and exaggerate its deadly effects with the aim of promoting expensive vaccinations. And it went without saying (of course) that the mainstream media and big tech companies were also in on the racket and busily suppressing any critical reporting of it.

Like most conspiracists, Junior was big on social media, but then in 2021 his Instagram account was removed for “repeatedly sharing debunked claims about the coronavirus or vaccines”, and in August last year his anti-vaccination Children’s Health Defense group was removed by Facebook and Instagram on the grounds that it had repeatedly violated Meta’s medical-misinformation policies.

But guess what? On 4 June, Instagram rescinded Junior’s suspension, enabling him to continue beaming his baloney, without let or hindrance, to his 867,000 followers. How come? Because he announced that he’s running against Joe Biden for the Democratic nomination and Meta, Instagram’s parent, has a policy that users should be able to engage with posts from “political leaders”. “As he is now an active candidate for president of the United States,” it said, “we have restored access to Robert F Kennedy Jr’s Instagram account.”

Which naturally is also why the company allowed Donald Trump back on to its platform. So in addition to anti-vax propaganda, American voters can also look forward in 2024 to a flood of denialism about the validity of the 2020 election on their social media feeds as Republican acolytes of Trump stand for election and get a free pass from Meta and co.

All of which led technology journalist Casey Newton, an astute observer of these things, to advance an interesting hypothesis last week about what’s happening. We may, he said, have passed “peak trust and safety”. Translation: we may have passed the point where tech platforms stopped caring about moderating what happens on their platforms. From now on, (almost) anything goes.

If that’s true, then we have reached the most pivotal moment in the evolution of the tech industry since 1996. That was the year when two US legislators inserted a short clause – section 230 – into the Communications Decency Act that was then going through Congress. In 26 words, the clause guaranteed immunity for online computer services with respect to third-party content generated by its users. It basically meant that if you ran an online service on which people could post whatever they liked, you bore no legal liability for any of the bad stuff that could happen as a result of those publications.

On the basis of that keep-out-of-jail card, corporations such as Google, Meta and Twitter prospered mightily for years. Bad stuff did indeed happen, but no legal shadow fell on the owners of the platforms on which it was hosted. Of course it often led to bad publicity – but that was ameliorated or avoided by recruiting large numbers of (overseas and poorly paid) moderators, whose job was to ensure that the foul things posted online did not sully the feeds of delicate and fastidious users in the global north.

But moderation is difficult and often traumatising work. And, given the scale of the problem, keeping social media clean is an impossible, sisyphean task. The companies employ many thousands of moderators across the globe, but they can’t keep up with the deluge. For a time, these businesses argued that artificial intelligence (meaning machine-learning technology) would enable them to get on top of it. But the AI that can outwit the ingenuity of the bad actors who lurk in the depths of the internet has yet to be invented.

And, more significantly perhaps, times have suddenly become harder for tech companies. The big ones are still very profitable, but that’s partly because they been shedding jobs at a phenomenal rate. And many of those who have been made redundant worked in areas such as moderation, or what the industry came to call “trust and safety”. After all, if there’s no legal liability for the bad stuff that gets through whatever filters there are, why keep these worthy custodians on board?

Which is why democracies will eventually have to contemplate what was hitherto unthinkable: rethink section 230 and its overseas replications and make platforms legally liable for the harms that they enable. And send Junior back to the soapbox he deserves.

What I’ve been reading

Here’s looking at us
Techno-Narcissism is Scott Galloway’s compelling blogpost on his No Mercy / No Malice site about the nauseating hypocrisy of the AI bros.

Ode to Joyce
The Paris Review website has the text of novelist Sally Rooney’s 2022 TS Eliot lecture, Misreading Ulysses.

Man of letters
Remembering Robert Gottlieb, Editor Extraordinaire is a lovely New Yorker piece by David Remnick on one of his predecessors, who has just died.