Happy Thanksgiving Eve! I know that I promised a full week off so that we could all focus on our pie consumption, but the news gods had other plans. So here we are, with a special edition of Nightcap to take a look at the absolutely bonkers business stories that have played out over the past several days. Buckle up! By Allison Morrow | | | | Last updated November 22 at 4:59 PM ET | |
|
| (Justin Sullivan/Getty Images) | What started as a juicy boardroom drama last Friday has, in just five days, become a saga that someone will almost certainly write a book and/or limited TV series about. A quick recap: - The board of OpenAI fired its CEO, Sam Altman, on Friday. Everyone was shocked. This is the guy leading the company behind ChatGPT, the most significant technology to come out of Silicon Valley in decades.
- The board stood behind their decision and hired another CEO. But it was also quietly second-guessing the move and trying to figure out a way to keep Altman.
- By Sunday, OpenAI's biggest shareholder, Microsoft, said it would hire Altman to head up a new in-house AI team.
- OpenAI's staff, meanwhile, told the board that if they don't get Altman back and resign themselves, virtually the entire 700+ team would quit.
- Late Tuesday, Altman was reinstated as OpenAI's CEO.
In a story with many wild moments, the wildest one, for me, is that we still don't have a clear sense of why Altman was fired in the first place. When the news came out Friday, OpenAI claimed that Altman had not been "consistently candid in his communications" with the board. Um. What? Rumors ran wild. While we can't say exactly why he was fired, we do know there were serious tensions between Altman and members of the board over how aggressively to push OpenAI's technology into the world. The board wanted to tap the brakes, and Altman seemed more willing to roll the dice (and in so doing, make a tidy profit). Why it matters OpenAI's GPT tech is the industry gold standard, and it is powerful stuff. The promise and the risk of AI are difficult to fully envision because of how new the tech is. At a basic level, the promise of AI is that it can make everything run smoother, automate mundane tasks, make the internet smarter and thus more helpful. ChatGPT can write your emails, tell you how to cook a turkey, translate text into just about any language, help you weigh pros and cons for a big decision, and much, much more. Of course, eventually that could mean ChatGPT can write pithy newsletters and help them navigate legal issues and do their accounting for them — all of which would put the humans in those professions out of work. And because AI systems are built by humans, who are flawed, and learn from data created by humans, the systems themselves can engrain bias. Already, police departments have used AI for facial recognition and even "predictive policing" — using computer systems to analyze large data sets to help determine where and when crimes are likely to be committed — which can exacerbate racial and socioeconomic biases. This week, I spoke with Jason Schloetzer, a professor at Georgetown's McDonough School of Business and an expert on AI's impact on the future of work. I wanted to know what kept him up at night when thinking about AI, and his answer has stuck with me: "What are our kids' kids' kids going to think about this … Was it much ado about nothing? Or did it ruin their lives?" Will they, he wondered, "have a "more productive, fulfilling, valuable life because of the pace of artificial intelligence discoveries, or is their life going to be ruined or drastically different because of the accelerating pace of AI discoveries? I think that's the philosophical tension. And now, we see it laid bare with the situation at OpenAI." | |
| Just when the crypto world thought it could move on from the Sam Bankman-Fried/FTX debacle, a fresh federal indictment comes down for the founder of another exchange. ICYMI: The billionaire founder of Binance, Changpeng Zhao, resigned as CEO and pleaded guilty to federal money laundering charges on Tuesday. Zhao, who goes by CZ online, agreed to plead guilty as part of a coordinated settlement in which Binance agreed to pay more than $4 billion in fines and other penalties. Zhao himself will pay $200 million in fines. He faces a maximum of 10 years in prison, though his ultimate sentence will likely be far lower. Why it matters There are few figures in crypto that have loomed quite as large in the past few years than CZ. Not only is Binance the world's biggest crypto exchange, it is orders of magnitude larger than its rivals. Up until recently, Binance boasted nearly 60% of the market share for crypto spot trading. Even as that share has waned to closer to 40% since US regulators amped up pressure on the company starting in June, no other exchange even comes close. The second and third largest exchanges each claim about 5%. But Binance's future at the top of the crypto world is now far from certain. Following a multiyear investigation, US authorities said Binance allowed bad actors on the platform, enabling transactions linked to child sex abuse, narcotics and terrorist financing. Further, Binance did not have protocols to flag or report transactions for money laundering risks, according to the Justice Department, and employees were well aware that such an oversight would invite criminals to the platform. As one compliance staffer wrote, according to court documents: "we need a banner 'is washing drug money too hard these days - come to binance we got cake for you.'" Too big to fail? The $4.3 billion settlement is one of the largest corporate penalties in US history. But it's unlikely to bankrupt Binance, and that is likely by design. Given its size, US authorities wouldn't have much incentive to fine Binance into oblivion. It is systemically important in crypto, aka too big to fail. "What this plea deal does is give Binance as a chance to live another day," said Yesha Yadav a law professor at Vanderbilt University and an expert on financial regulation. "I think that reflects a worry that if Binance were to be killed, that would cause further damage to average folks who hold money on it to the industry as a whole." | |
| X, formerly known as Twitter, is spiraling even faster than I expected when Elon Musk bought it just over a year ago. ICYMI: It started with an antisemitic tweet, which I wrote about last week here. Unsurprisingly, many advertisers on X were appalled and began rushing the exits. IBM was one of the first, suspending its advertising on X after a report from progressive media watchdog Media Matters found its ad had run alongside pro-Nazi content on the platform. A bunch of media brands followed, including Disney, Paramount, Comcast, Lionsgate, NBCUniversal and Warner Bros. Discovery (CNN's parent company), though they did not specify their reasons for halting advertising on X. Apple also reportedly pulled its ad spend on X as of last week. My colleague Clare Duffy writes: The exodus of advertisers marks a deepening crisis for X, which was already struggling to woo brands back to the platform following Musk's takeover last year. And it comes as growing numbers of X users also decamp to alternate platforms. The White House joined Meta's Threads this week, giving a boost to the Twitter-like rival. Meanwhile, the pro-Nazi accounts identified by Media Matters remain active on the site, along with other far-right and White supremacist accounts. Musk himself moved on to boosting the several-years-old, widely debunked "Pizzagate" conspiracy theory that in 2016 prompted a gun-wielding man to fire shots inside a DC pizza shop. Of course, some advertisers have remained on X despite the hate speech. The National Football League — one of the platform's biggest ad partners — told CNN Tuesday that it had repeatedly expressed concerns about hate speech to X leadership directly, though it has not pulled its spending from the platform. Others still running ads on X include Walmart, State Farm, Wendy's, Office Depot, The New York Times, The Washington Post, The Economist, USAA insurance, Formula 1 and Mondelēz International. | |
| Last updated November 22 at 4:00 PM ET | | |
| |
Comments
Post a Comment