簡介
Presenters：Kelly Yu and Raj Shroff
RTHK's podcast on the latest tech news
最新
Meta's new watermark tool to fight deepfakes, YouTube tests a feature that chooses what you see next
As AI continues to improve, it's getting harder to tell when videos are real, or generated by a computer.
Tech giant Meta says it wants to help clear up the confusion with a new tool that basically adds an invisible watermark to AI-generated videos.
You can't actually see these watermarks, it's a digital signature that's added on to AI-videos and is designed to be hard to remove.
So, will this make it easier for regular people to tell if a video is real or fake?
Listen to what AI consultant Raj Shroff has to say in this week's Tech Tuesday.
You'll also hear about a new 'Play Something' button on YouTube that basically decides what you watch next, and how this could affect how we browse content.
重溫
Zoom’s new AI avatar may revolutionise workplace, controversy over LinkedIn's new data collection policy
Presenters：Kelly Yu and Raj Shroff
Office workers will soon be able to talk to colleagues using digital lookalikes in their Zoom workspace, thanks to a new AI avatar feature announced by the video conferencing platform.
To use this tool, the user will first have to record a video clip. Zoom’s AI will then generate an avatar that looks and sounds like him or her.
Set to launch in the first half of 2025, this feature enables the digital clone to deliver pre-scripted messages.
While some believe it will save time and boost productivity, concerns have been raised about potential misuse for scams and deepfakes.
So is there a way to strike a balance, and how might this shape the future of remote work? Tune in to AI consultant Raj Shroff’s take.
You’ll also hear how an update to LinkedIn’s data collection policy for AI model training sparked privacy concerns.
