Welcome to the newest edition of FindBiometrics’ AI update. Here’s the latest big news on the shifting landscape of AI and identity technology:
Google has launched Gemini, the flagship version of its ChatGPT rival. The name reflects a rebrand from Bard, the previous moniker of its chatbot. Like OpenAI, Google is offering a free basic version of the AI tool as well as a more advanced subscription version called Gemini Advanced.
The chipmaker Arm has beaten analysts’ expectations with Q4 revenues of $824 million, thanks to what CEO Rene Haas called the “profound opportunity” of demand for AI applications. Nvidia’s Grace Hopper, Microsoft’s Cobalt and Amazon’s Graviton chips are all based on Arm’s designs.
Zededa has raised $72 million in a new funding round, bringing its valuation to about $400 million. The startup makes edge computing tools for the enterprise, and is seeing surging interest thanks to AI, which is replacing the traditional analytics software that’s normally used in edge computing.
The Department of Homeland Security is setting up a new “AI Corps” and is looking to recruit individuals with expertise in biometrics, among other areas. AI-related skills listed in some of the new department’s job postings include those involving facial recognition, machine vision, and natural language processing. The DHS AI Corps is looking to hire 50 experts over the next year.
Member states of the European Union have reached a unanimous agreement on the planned Artificial Intelligence Act, with deputy ambassadors formally approving a compromise text that was reached in December. Next steps for the proposed legislation include a committee vote this month, followed by a plenary vote that will likely occur in April.
A deepfake-driven scam has cost a multinational company HK$200 million, or about $25.6 million, after a synthetic copy of its Chief Financial Officer ordered money transfers in a video call with a Hong Kong employee. Deepfakes of multiple individuals were reportedly on the video call, with the victim of the scam being the only genuine employee present. The employee realized it was a scam after making further inquiries with the company’s senior leadership.
Meta’s independently-run Oversight Board has called Meta’s policy concerning manipulated videos “incoherent” in a new ruling on a doctored video depicting President Joe Biden. The board said that Meta’s policy should cover both video and audio content, and should apply labels to manipulated media rather than removing any such posts from the Facebook platform. Facebook told the Oversight Board that it is working to update its policies to better “respond to the evolution of new and increasingly realistic AI.”
The Federal Communications Commission has announced a ban on robocalls that feature AI-generated voices, framing them as a violation of the 1991 Telephone Consumer Protection Act. “Bad actors are using AI-generated voices in unsolicited robocalls to extort vulnerable family members, imitate celebrities, and misinform voters,” explained FCC chairwoman Jessica Rosenworcel.
On that note, New Hampshire’s Attorney General has announced a criminal investigation into a Texas company alleged to have been making robocalls impersonating President Joe Biden. The calls told thousands of New Hampshire voters not to vote in the state primary, saying they should instead “save your vote for the November election.” The AG is issuing subpoenas to Life Corp., to Lingo Telecom which allegedly carried its calls, and to others thought to have been involved.
The chatbot’s take: This week, we decided to give Gemini a spin.
February 8, 2024 – by Alex Perala