Zoom has changed its terms of service in response to criticism regarding the possibility of their AI models being trained using customer calls.
The firm emphasized in a blog post that AI had not been utilized for audio, video, and chat purposes without prior consent.
After users detected modifications to the company's terms of service in March, which gave rise to fears that Artificial Intelligence might be trained, the video-calling app took action.
The firm declared that it implemented the modifications to increase transparency.
In June, Zoom rolled out new AI-powered features, including one that allows customers to summarise meetings without needing to record the full session. The features were available on a complimentary trial.
However, certain specialists cautioned that the initial language of the service's terms could have allowed Zoom to obtain more user information than necessary, even from conversations taking place through their platform.
Speaking with the BBC ahead of their updates to the conditions of service, data security professional Robert Bateman declared: "The conditions seemed to give the service provider considerable liberty to make use of data produced by its clients for different goals."
He indicated that there was an uncertainty with respect to potential risks, adding that "one should be wary when encountering expansive contractual stipulations such as these".
On Monday, Zoom issued a revision of its terms which included the provision that the company "will not use audio, video or chat customer content to train its artificial intelligence models without gaining the consent of the customer".
Computer programmes or tools powered by AI are able to execute tasks that are usually carried out by individuals. They are trained using extensive data and algorithms to imitate behaviour that is similar to a human's.
The use of the internet to collect data for use in creating AI applications has caused alarm, leading to legal action over the prospect that private information, confidential information, or copyrighted material could be included in the collections.
Zoom has given increased attention to products using AI this year, as is the case with many other tech firms, due to the current trend of enthusiasm surrounding AI.
The Open Rights Group, an advocate for digital privacy, has voiced concern over Zoom's decision to make these new features available as a free trial and appeal to customers to 'opt in', which is seen as particularly troubling.
Abby Burke, the policy manager for data protection, informed the BBC that, prior to the most recent amendment to the Zoom privacy policy, their statement that customers were to be asked for permission to use their data in order to train AI models was not clear.
A Zoom representative once again affirmed on Monday that customers have the choice to activate generative AI characteristics, as well as to independently decide whether to provide customer data to Zoom for "product advancement objectives".
Pictures that were featured in Monday's blog provided evidence of cautionary alerts for participants of gatherings that utilize AI-based technologies, prompting them to either accept the directive to be trained or to leave the meeting.
Smita Hashim, Zoom's chief product officer, noted that account owners and administrators have the choice of turning on features that are still being tested out, and that those who activate them will "be greeted with a transparent consent process when utilizing customer content in order to train our AI models".
Comentarios