Online sites relying on AI as more staff switch to working from home

Artificial Intelligence is taking the strain at several high-technology companies as employees switch to homeworking because of movement restrictions triggered by the fight against novel coronavirus.

Major websites, including Google, Facebook, Twitter, and YouTube, are reportedly relying on AI and other automated tools most heavily to find and remove inappropriate content on their platforms.

The BBC reports that the tech giants have admitted their additional reliance on AI could lead to some mistakes being made in the removal of harmful content.

The companies are especially keen to remove fake news related to the virus and the COVID-19 respiratory disease it causes, the broadcaster said.

Twitter issued a statement saying the machine-learning apparatus it is now relying on and the increased automation of content screening may "sometimes lack the context that our teams bring, and this may result in us making mistakes".

The company said, as a result of the fallibility of AI systems, it will not permanently ban any users because of removed posts.

Facebook said in a blogpost: "With fewer people available for human review, we'll continue to prioritize imminent harm and increase our reliance on proactive detections in other areas."

Google, which owns Youtube, ordered its entire worldwide full-time workforce to work from home because of the pandemic.

"This means automated systems will start removing some content without human review," YouTube said in a blog. "As we do this, users and creators may see increased video removals, including some videos that may not violate policies."

The company said all users will be able to appeal against decisions made by the automated systems and have them reversed if they are found to be wrong at a later date, although the company admitted the appeals process will take longer than usual.

The United Kingdom's government has spoken of the importance of countering misinformation during the novel coronavirus outbreak, although the nation's cross-party Digital, Culture, Media and Sport committee has questioned why Downing Street took two months to set up a special unit dedicated to countering the spread of disinformation.

Technology companies are also looking to use AI on the frontline in the battle to understand the virus and develop treatment methods.

Business Insider magazine, the United States online news site, says Microsoft Research, the Allen Institute for AI, and the Chan Zuckerberg Initiative have teamed up to establish an online database of 29,000 scientific articles about the disease.

They have stored the articles in machine-readable form and are inviting AI experts to scan them for free, to look for answers.

"We are issuing a call to action to the word's artificial intelligence experts to develop text－and data-mining tools that can help the medical community develop answers to high-priority scientific questions," the partners said in a statement while launching the COVID-19 Open Research Dataset.