| Hi, it's Davey in New York. While Elon Musk says he was playing for laughs when he circulated a deepfake of a campaign ad for Vice President Kamala Harris, misinformation experts say there's a serious issue being missed. But first... Three things you need to know today: • Meta shares jump on the company's strong ad growth bolstered by AI
• Nvidia CEO Jensen Huang adds nearly $12 billion to his fortune in the chipmaker's rally
• NBC is betting on social media influencers to boost Olympics coverage In the past week, Elon Musk caused a stir on X — again. The billionaire owner of the platform shared an AI "parody" of a Kamala Harris campaign ad, featuring a voice similar to the presumptive Democratic nominee, describing herself as the "ultimate diversity hire" and "a deep state puppet." The narration was clearly generated by artificial intelligence, but Musk didn't disclose this, triggering a wave of backlash. It took Musk another three days to say anything else on the matter, time that could have been used for earnest reflection on the proliferation of AI tools that has caused the mainstreaming of synthetic media to happen at lightning speed. Instead, Musk resurfaced with a crude joke referencing an internet meme about male genitals. The attention span on Musk's action may have been short-lived, but the way he behaves online — as a major social media influencer who can point his more than 192 million followers at an idea just by posting it — still matters. If only it was more surprising. As my colleagues at Bloomberg have reported, this is just how Musk typically reacts. On the fifth day of the billionaire's takeover of Twitter, he composed a post in front of his lieutenants: "If I had a dollar for every time someone asked me if Trump is coming back on this platform, Twitter would be minting money!" As soon as one ad exec strenuously objected, Musk laughed and posted the tweet anyway — then fired the dissenter soon after. "Humor" and "parody" have long been the defenses he has mounted in the face of criticism. "Common sense would suggest that individuals with the greatest influence online should be held to higher standards of platform policy compliance and basic human decency," Nora Benavidez, the senior counsel at Free Press, a digital civil rights group, told me when I asked for her thoughts on the affair. "But having bought his own platform, Musk thinks he can bend the rules and basic decency however he wishes." The incident could have been washed away in a news cycle saturated with so many other topics worthy of focus — who Harris' VP pick will be, Donald Trump's questioning of her racial identity, the Olympics! — but it's noteworthy as the public wades through which influential person's views to take seriously and how to parse them in the midst of a presidential campaign. "Humor," said Laura Edelson, an expert on online misinformation with Northeastern University, "is how fringe ideas get normalized. It allows people who want to do things that are outside the normal boundaries of society to deflect criticism." Musk's actions show he isn't thinking about the overall health of the information ecosystem, she said. "He's a partisan actor — which he's allowed to be — who wants to score political points and doesn't care as much about how he does that," Edelson wrote in an email. And it's true that Musk knows his audience and knows that his persona on X gets engagement — you can't argue with the 245,000 retweets and over 133 million views on his original post. For Emerson Brooking, a resident senior fellow at the Atlantic Council who studies online networks, Musk's stewardship of X has already harmed the state of US online discourse. "He may now meaningfully influence the outcome of an election where he is actively promoting one candidate and sharing false content about another," Brooking said in an interview. "But there's no real plan here. In many ways, that makes it sadder." What has been sidelined by the episode is a considered discussion of whether the use of deepfakes in politics is acceptable. Instead, Musk's antics get the attention. "We don't have bipartisan agreement on deepfakes," Edelson said. "This is an unstable equilibrium. If we can't come to an agreement that neither side will use these tactics, eventually both sides will."—Davey Alba Please note: Our email domain is changing, which means you'll be receiving this newsletter from noreply@news.bloomberg.com. Update your contacts to ensure you continue receiving it -- check out the bottom of this email for more details. Google is cracking down on deepfake porn in search results. When AI-generated adult content features a real person without their permission, that person can request its removal from search results. Now, when Google decides a takedown is warranted, it will filter all explicit results on similar searches and remove duplicate images. The company also said it has improved its search ranking systems so that explicit fake content won't appear in top results. The US may restrict China's access to AI memory chips in the latest escalation in the Biden administration's effort to curtail the export of advanced technology. Arm shares tumble after the CEO reported the chip designer has seen signs of market weaknesses. Qualcomm gives a bullish revenue forecast, suggesting a smartphone rebound is ahead. | 
No comments:
Post a Comment