Oracle has begun auditing the algorithms of social media giant TikTok in order to ensure they are not being manipulated by Chinese authorities.
As reported by Axios, (opens in new tab) the US cloud giant has already started examining both TikTok’s algorithms and content moderation models following long-running concerns about collusion between the Chinese government and TikTok parent company ByteDance.
It comes shortly after TikTok migrated American users’ data to servers owned by Oracle US in an attempt to lessen worries about possible snooping and manipulation.
TikTok scrutiny
A TikTok spokesperson told Axios that the new partnership would allow Oracle to carry out, “regular vetting and validation” of the company’s content recommendation and moderation models.
Past reports had claimed that TikTok blocks or marks down certain videos critical of the Chinese government, as well as censoring videos that mention subjects such as Tibetan independence or the Tiananmen Square massacre.
The reviews should mean Oracle can examine how TikTok’s algorithms surface content “to ensure that outcomes are in line with expectations and that the models have not been manipulated in any way,” the spokesperson added.
Providing Oracle with such visibility, “will ensure that content continues to be flagged and actioned appropriately based on our Community Guidelines and no other factors.”
READ MORE:
The move is the latest in a line of TikTok attempts to help reassure the US government over fears that user data was being manipulated.
The Trump administration had taken a hard line against TikTok and ByteDance, with a 2020 ruling by a US national security panel ordering the company to divest its US operations, alleging that the data of US users could find its way into the hands of the Chinese government.
However, the Biden administration has looked to work with the company, which has launched Project Texas, a wide-ranging attempt to persuade US lawmakers that user data is safe.
Along with moving US user data to Oracle’s facilities, TikTok has also separated its US operation backend functions and code as part of Project Texas.