In the second season of BBC mystery thriller The Capture, deepfakes threaten the long run of democracy and UK national security. In a dystopia set in present day London, hackers use AI to insert these highly realistic false images and videos of individuals into live news broadcasts to destroy the careers of politicians.

But my team’s research has shown how difficult it’s to create convincing deepfakes in point of fact. In fact, technology and artistic professionals have began collaborating on solutions to assist people spot bogus videos of politicians and celebrities. We stand an honest likelihood of staying one step ahead of fraudsters.

In my research project, Virtual Maggie, I attempted to make use of deepfakes to digitally resurrect former UK prime minister Margaret Thatcher for a brand new drama. After months of labor, we were unable to create a virtual Maggie acceptable for broadcast.

Producing convincing deepfakes in high definition requires top spec hardware, plenty of computer time, and human intervention to repair glitches within the output. This didn’t stop me having fun with The Capture, despite knowing that Ben Chanan’s drama was not a scenario more likely to play out within the near future. Like every good dystopia, it had the seed of something that may sooner or later be possible.

The use of deepfakes since they began in 2017 has been shocking. The majority of deepfakes on the web are assaults on women, grabbing facial images without consent and inserting them into pornographic content. Deepfakes expert Henry Ajder found that 96% of deepfakes found online were pornographic, and 100% of those were video images of girls.

The premise of The Capture is grounded in facts. Deepfakes threaten democracy. In the 2019 UK general election, artist Bill Posters released a provocative video of Boris Johnson saying we must always vote for Jeremy Corbyn.

Posters’ deepfake was rather a lot more convincing than the glitchy Russian deepfake showing Ukraine President Volodymyr Zelenskyy asking his troops to give up. Yet, unlike the Kremlin, the British artist made it obvious his AI Boris was unreal by having “Boris” direct viewers to an internet site about deepfakes. He aimed to focus on our vulnerability to faked political propaganda.

Deepfakes may not yet be typically convincing enough to idiot people. But creative work typically involves an unwritten agreement between creator and viewers to suspend their disbelief.

A fork within the road

The threat from deepfakes has led to an intensive seek for tech solutions. A coalition of firms has formed the Content Authenticity Initiative (CAI) to offer “a approach to evaluate truth within the media presented to us”.

It’s a promising approach. CAI collaborators and technology firms Truepic and Qualcomm have created a system that embeds the history of a picture in its metadata so it may possibly be verified. US photographer Sara Naomi Lewkowicz has accomplished an experimental project with CAI that embeds source information in her photos.

But creative and technology professionals don’t necessarily wish to trammel the emerging technology of deepfakes. Researchers on the Massachusetts Institute of Technology Media Lab have been brainstorming ways of putting deepfakes to good use. Some of those are in healthcare and treatment.

Research engineers Kate Glazko and Yiwei Zheng are using deepfakes to assist individuals with aphantasia, the lack to create mental images in your mind. The breakup simulator, under development, goals to make use of deepfakes to “alleviate the anxiety of inauspicious conversation through rehearsal”.

The most profound positive uses for deepfakes include campaigns for political change. The parents of Joaquin Oliver, killed in a highschool shooting in Florida in 2018, used the technology to bring him back in a forceful video calling for gun control.

Getting creative

There are also cultural applications of deepfakes. At the Dali Museum in Florida, a deepfake Salvador Dali welcomes visitors, telling them about himself and his art. Researcher Mihaela Mihailova says this offers visitors “a way of immediacy, closeness, and personalization”. Deepfake Dali even offers you the prospect to take a selfie with him.

Deepfakes and AI-generated characters may be educational. In Shanghai, during lockdown, Associate Professor Jiang Fei noticed his students’ attention dropped during online lessons. To help them focus higher he used an anime version of himself to front his teaching. Jiang Fei said: “The enthusiasm of the scholars in school, and the advance of the standard of homework have made obvious progress.”

Channel Four used its 2020 alternative Christmas message to entertain viewers with a deepfaked queen, while making a serious point about not trusting the whole lot we see on video.

A growing network of film producers, researchers and AI technologists within the UK, hosted by the University of Reading and funded by the Alan Turing Institute, is searching for to harness the positive potential of deepfakes in creative screen production. Filmmaker Benjamin Field told the group during a workshop how he used deepfakes to “resurrect” the animator who created Thunderbirds for Gerry Anderson: A Life Uncharted, a documentary in regards to the troubled lifetime of the youngsters’ TV hero.

Field and his co-producer, Anderson’s youngest son Jamie, discovered old audio tapes and used deepfakes to construct a “filmed” interview with the famous puppeteer. Field is amongst a small group of creatives determined to seek out positives ways of using deepfakes in broadcasting.

Deepfakes and AI characters are a part of our future and the above examples show how this may very well be at the very least partly positive. But we also need laws to guard people whose images are stolen or abused, and ethical guidelines on how deepfakes are utilized by filmmakers. Responsible producers have already formed a partnership on AI, and drafted a code of conduct which could help avert the catastrophic vision of the long run that we saw within the The Capture.

This article was originally published at