Welcome to the newest edition of ID Tech’s AI Update. Here’s the latest big news on the shifting landscape of AI and identity technology:
Disney has joined in on the AI sector’s circular dealmaking, partnering with OpenAI in an agreement that will see the latter license Disney IP for use in its Sora video generation platform. Disney, meanwhile, is investing $1 billion in OpenAI. Effectively, the agreement means that more than 200 Disney characters will be able to appear in Sora-generated videos. It offers serious validation to the idea that creative IP holders see opportunities in the world of generative AI, and not simply copyright threats.
The White House has issued an Executive Order aimed at discouraging state-level AI regulation. Titled “Ensuring a National Policy Framework for Artificial Intelligence”, the order instructs the Commerce Department to draw up a list of state-level AI laws that it deems onerous, and it instructs the Attorney General to set up a dedicated task force whose sole responsibility is to challenge such state laws. It also adds some levers in the form of withholding certain grants and funding from states with heavy AI regulation.
The Chinese government is reportedly considering offering AI chip subsidies in the range of 200 billion to 500 billion yuan (about $28 billion to $70 billion), offering the latest sign of its efforts to ramp up domestic production capabilities. At the same time, China’s Ministry of Industry and Information Technology (MIIT) has added domestic AI processors to an official government procurement list for the first time, a move designed to push ministries, state-owned enterprises, and public institutions to prioritize Chinese-made AI accelerators over foreign options. This comes shortly after the US government gave formal approval for the sale of Nvidia’s H200 chips to China, under certain conditions.
Meanwhile, the Silicon Valley penchant for AI talent poaching has been exported to China, where Tencent has been aggressively recruiting AI researchers from ByteDance as it tries to accelerate its AI push, including offers that can reach roughly double a researcher’s pay. According to The Information, Tencent is also reorganizing its AI org and demoting some non-research leaders, a signal that its leadership believe it is bottlenecked by research velocity, not product distribution.
The European Commission has launched an antitrust investigation into Google over a couple of AI concerns. First, the EC wants to determine whether Google is effectively forcing publishers to opt into its AI search services, since if they don’t opt in they are less likely to show up in search results. It is also investigating Google’s potentially privileged access to YouTube for training its AI algorithms, with other AI firms being potentially shut out. As in last week’s EC investigation announcement into Meta and its WhatsApp AI services, this latest investigation also seems to hinge on the idea that a tech company shouldn’t be able to get an advantage from its own services in the AI race.
Semiconductor maker Broadcom has revealed that Anthropic is its mystery customer that committed to ordering over $10 billion worth of chips earlier this year. Broadcom’s management had brought the huge order to light in early September without disclosing the name of the customer, leading many analysts to speculate that it might have been OpenAI, given its access to huge amounts of capital. But on an earnings call this week, Broadcom CEO Hock Tan revealed that it was actually OpenAI’s arch-rival, and disclosed that Anthropic had made an additional, $11 billion commitment. Nevertheless, Broadcom’s stock slid after the call, on which Tan also acknowledged that gross margin will probably shrink next quarter due to a stronger focus on lower-margin offerings.
Europol and affiliated police authorities have disrupted a criminal network that used deepfakes to rake in millions through fraud. The police say that the sprawling fraud network wrought about €700 million in damages, and that much of it revolved around tricking people into investment schemes through crypto networks. “In recent years, deceptive advertisements impersonating renowned media outlets, celebrities and politicians – often using deepfake videos – have posed a significant global challenge,” Europol stated in a press release.
The chatbot’s take: We asked for some context about Disney’s dealmaking habits.
–
By Alex Perala
Follow Us