AI has been a topic that has been popping up on most people’s news feeds recently, but not always for the best reasons. The Federal Communications Commission (FCC) recently made it illegal to use AI-generated voices for robocalls in the U.S., an extension of the Telephone Consumer Protection Act aimed at protecting consumers from unsolicited calls. This decision follows incidents of AI-generated voice calls being used maliciously, including fake messages mimicking U.S. President Joe Biden.
Meanwhile, OpenAI is rumored to be developing an autonomous AI assistant that could greatly enhance how users interact with devices by performing tasks directly, although this raises some important privacy and security concerns. Microsoft is also expanding its AI presence in India through a partnership with Sarvam AI, in an effort to integrate Sarvam&'s Indic voice large language model into Microsoft&'s Azure AI infrastructure.
FCC Cracks Down on AI-Generated Voice Scams
The Federal Communications Commission (FCC) declared that the use of AI-generated voices in robocalls is now officially illegal in the United States. This decision comes as an extension of the Telephone Consumer Protection Act (TCPA), which is aimed at protecting consumers from unsolicited calls, which already prohibits robocall scams. The FCC&'s unanimous adoption of a Declaratory Ruling on Feb. 8 classifies calls made with AI-generated voices as &'artificial&' under the TCPA, thus providing state attorneys general with the capabilities to prosecute those behind these invasive calls.
This crackdown on AI-generated voice calls follows incidents where this technology was used for malicious purposes, including a recent case in New Hampshire where residents received fake voice messages, mimicking U.S. President Joe Biden, advising against voting in the state’s primary election. These kinds of actions not only exploit vulnerable people but they certainly also pose a big threat to the integrity of electoral processes and personal security.
FCC chair Jessica Rosenworcel emphasized the urgency of addressing this issue, pointing out the use of AI to extort, impersonate, and disseminate false information. With the advancement of technology, AI-generated voices have become extremely sophisticated, capable of imitating celebrities, political figures, and even close family members.
The FCC&'s new rule, effective immediately, reinforces the existing framework of the TCPA, which mandates telemarketers to get written consent from people before initiating robocalls. The ruling is a big step forward in making sure that AI-generated voices are subject to the same regulatory standards, thereby tightening the leash quite a bit on unauthorized and deceptive telemarketing practices.
The alleged perpetrator behind the deceptive Biden robocalls in New Hampshire has been traced back to a Texas-based firm, Life Corporation, and Walter Monk. The Election Law Unit has since issued a cease-and-desist order to the implicated people.
OpenAI&'s Vision for Autonomous Assistants
Meanwhile, OpenAI is reportedly developing an autonomous AI assistant that could revolutionize the way users interact with their devices by performing tasks directly. This new product could build on the capabilities of generative AI systems like ChatGPT and Google’s Gemini, which produce human-like media but lack the ability to execute real-world actions without external applications. The envisioned technology would potentially beat current smart assistants by leveraging large language models for more autonomous operations.
The speculated capabilities of this AI system include executing tasks on a user&'s device, like data management and interacting with interfaces in ways previously only limited to humans, like solving CAPTCHA puzzles.
However, the development of an autonomous assistant like this naturally raises some serious privacy and security concerns. Current generative AI systems rely on cloud computing, suggesting that a fully autonomous assistant might require a lot of data access and processing capabilities, posing privacy risks. Furthermore, giving an AI system extensive device privileges could also introduce new cyberthreats as the AI is allowed to operate with high levels of autonomy.
As OpenAI has yet to comment about the rumors floating around in the crypto and AI communities, the full extent of the project&'s ambitions and its approach to navigating these challenges still only remains speculative. Nonetheless, the concept of an autonomous AI assistant suggests a future where the interaction between humans and technology becomes even more seamless, although there are still a few hurdles to overcome.
Microsoft Partners with Sarvam AI to Boost AI Innovation in India
India has its own plans for AI. Microsoft is making big strides in expanding its AI footprint in Asia through a strategic partnership with Indian startup Sarvam AI. This collaboration aims to integrate Sarvam&'s advanced Indic voice large language model (LLM) into Microsoft&'s Azure AI infrastructure, leveraging Azure OpenAI and Machine Learning services to enhance the hosting, training, and scaling of Sarvam&'s LLMs. This initiative is part of Microsoft&'s broader commitment to foster AI-driven growth and innovation in India, aligning with its vision to support the country&'s transformation into an "AI-first nation."
Puneet Chandok, President of Microsoft India and South Asia, also emphasized the importance of this partnership in promoting homegrown innovation and ensuring that AI solutions are accessible to everyone, regardless of language or background. Sarvam&'s LLM, which initially focuses on the Hindi language, aims to provide a natural voice-based interface that could revolutionize the way applications interact with users by being more precise and cost-effective. Hindi, being the most widely spoken language in India, serves as a critical medium for making technology accessible to nearly 609 million speakers across the country.
This partnership not only paves the way for local businesses to adopt generative AI technologies quickly and responsibly but also reflects Microsoft&'s ongoing efforts to incorporate AI across its entire technology stack. The company&'s investment in AI has been a key driver of its growth, as seen in the impressive 18% increase in its year-end earnings for 2023 compared to 2022. This growth has contributed to Microsoft&'s valuation passing $3 trillion, overtaking Apple as the most valuable publicly traded company.
In addition to its partnership with Sarvam AI, Microsoft has also been actively forming alliances with other organizations, including a collaboration with telecom giant Vodafone and the American Federation of Labor and Congress of Industrial Organizations in the United States. These partnerships are aimed at developing more "worker-focused" AI solutions.