Today’s cache | Google’s ambitions in AI and the Web | Photo credit: REUTERS
(This article is part of Today’s Cache, The Hindu’s newsletter on emerging themes at the intersection of technology, innovation and politics. To receive it in your inbox, subscribe here.)
Google’s next AI project
Google is preparing to display a product named Project Jarvis and possibly the next iteration of the Gemini modelreported The informationciting anonymous sources. The Jarvis Project could potentially take control of a user’s web browser in order to help them accomplish tasks that would otherwise require several separate steps and processes. While AI models have so far been developed and integrated into the web experience to provide summaries and answer questions, the Jarvis project could go further. Others working on similar concepts include Google-backed Anthropic and Microsoft-backed OpenAI. Google has not yet confirmed the product or its release schedule.
However, a number of issues remain to be resolved, including the question of how AI-based web tools will deal with paywalls, as well as the question of how well they will be monetized with ads and sponsored entries.
OpenAI’s Whisper faces complaints
Although OpenAI has strongly endorsed its AI-powered transcription tool, commonly used in hospitals and call centers, users complain that Whisper makes up text that no one has actually spoken, which has potential repercussions. OpenAI has warned against using its tools in high-risk situations, but Whisper is still used in medical settings to help doctors record and review their patients’ information. However, computer scientists noted 187 hallucinations in more than 13,000 clear audio clips studied.
Hallucination is an ongoing challenge that AI scientists seek to solve, because models that hallucinate significantly increase the risk of making serious errors and biases when deployed in real life. However, experts are particularly concerned that the consequences of using Whisper in hospitals may not even be fully known due to medical confidentiality.
Microsoft lays off two employees
Microsoft fired two employees for holding a position vigil at its headquarters to mourn the Palestinians killed by Israel in Gaza. The lunchtime event took place on Microsoft’s campus in Redmond, Washington. However, Abdo Mohamed and Hossam Nasr were later fired. Both were also part of the No Azure for Apartheid employee coalition that opposes Microsoft’s sales of technology to the Israeli government as it carries out military action in Gaza, killing and displacing people. thousands of people. Microsoft confirmed that some layoffs had occurred, but cited its internal policies and privacy policy in declining to provide more information about the action.
In the United States, pro-Palestinian tech workers have come together in the past to call attention to how their companies are providing high-level technology to Israel despite the ongoing war. Google also fired employees following sit-ins against its relations with the Israeli government.
Published – October 28, 2024 at 11:40 AM IST
