Zoom introduced AI Companion 2.0 to enhance task management and organization. Users can create custom avatars for sending video messages to teams. Avatars will use AI to mimic users’ appearance and voice in videos.
In line with its broader AI initiatives, Zoom will soon introduce a feature that allows users to create a digital avatar of themselves for sending brief messages to their teams, as reported by The Verge. To create this digital representation, users must record an initial video, which Zoom’s AI will use to generate an avatar that resembles both their appearance and voice. Once set up, users can type out messages for their AI avatar to deliver, facilitating communication through Zoom’s Clips feature for sending quick video updates to colleagues, as noted by The Verge. Smita Hashim, Zoom’s chief product officer, explained that these custom avatars are designed to enable “asynchronous” communication among colleagues, making interactions “faster and more productive,” as reported by TechCrunch She emphasized that “avatars save users precious time and effort recording clips, and enable them to scale video creation,” as reported by TechCrunch. Zoom is also mindful of the potential risks associated with deepfakes. These avatars are being introduced at a time when deepfakes are becoming more common, making it harder to tell what is true and what is disinformation, as recently reported by the Insikt research group. In a press briefing, Hashim noted that the company is approaching this issue with caution by implementing advanced authentication methods, watermarking technology, and stringent usage policies, as reported by The Verge. However, it remains unclear how Zoom will specifically prevent malicious use of its tool for generating deceptive videos. A mock-up provided by the company displays a visible watermark in the upper-right corner of videos featuring custom avatars. Still, TechCrunch points out that such watermarks can easily be removed using screen-recording software.
