Arlene Foster: Looking at the murky world of online

India, the world’s largest democracy is currently in the middle of its general election.

Due to the vast size of the population, the election process will last for 44 days. India’s eligible voters at 968.6 million are more than twice the size of the whole European Union at 448 million.

Meanwhile, in the United States of America, we are all watching who the next President of the great super-power will be. That will be decided in November, and it is looking likely, according to most “informed” commentators that it will be November when the United Kingdom will also finally go to the polls.

How we consume our information on politics and news generally, has changed so much from when I first got involved in politics. Then if a significant speech were made to a large gathering of people, it would have been written up in the newspaper the following day.

Now, we get a blow-by-blow account of the speech and its key messages from social media “live” from the venue. I am not sure how Jim Molyneaux or indeed Margaret Thatcher would have survived in the world of X and Snapchat, but I do believe that before the era of 24-hour news, politics was a much more thoughtful and considered place; a time when policies meant more than the latest gossip online.

As Thatcher herself once said it is ideas which are the stuff of politics.

An astute man once said to me, that we had now access to much more information than ever before, but information is not wisdom, and it was the latter which was needed today as we did yesterday. I totally concur – we can be bombarded with information from lots of various sources but are we really any wiser?

And just as important is the source of the information – how can we trust what we see as being true?

It is important when we are making decisions about who to choose as our next Member of Parliament that we have the truth and not a fictionalised account of their record or beliefs – that applies whether we are here in the UK, India or in America where it is becoming nearly impossible to separate fact from fiction, especially online.

Since taking up my seat in the House of Lords I have spent some time looking at the online world and how it deals especially with vulnerable people – the young, those with disabilities, and the elderly. Since I started looking at this world, I have been discovering just how sophisticated the digital space is. One of those discoveries has been around the use of deepfakes.

Deepfakes are non-consensually Artificial Intelligence (AI) generated voices, images, or videos that a reasonable person would mistake as real. The deepfakes usually involve sexual imagery, fraud, and misinformation.

This may sound a bit far-fetched but I should remind readers that in the last NI Assembly election two female candidates, one DUP and the other SDLP were targeted by deepfakes. They both got re-elected regardless but for them both it was a traumatic experience and no doubt the deepfakes were generated to interfere with their chances of being elected.

Deepfakes pose serious threats to both personal liberty and global security and Governments across the world need to act.

We used to be told that the camera never lies, but regrettably, that is no longer true. The camera does lie as do videos which can be generated with audio and images that really do look like the real thing.

With this murky world online all around us, deepfakes are ripe for abuse as a means of manipulating elections – something I have raised on the floor of the House of Lords. So, what is to be done?

Some political deepfakes are unsubtle and easy to detect as such, others are nearly impossible to detect as being fakes.

One of the most impactful deepfakes so far in the UK was an audio clip which appeared to show Sadiq Khan the current mayor of London suggesting that the Remembrance weekend should be postponed in favour of a pro-Palestinian march.

Within a brief time, the clip went viral on X (formerly Twitter) and many protestors turned up near the Cenotaph in London to defend it, all based on a deepfake.

After the Online Safety Act, it is now illegal to circulate such information which one knows to be false and which one intends to cause harm, but it is still legal to create such content. In practice, the initial creator of the political misinformation is likely to remain anonymous and therefore evade enforcement in any event.

The government has now announced that they will ban the creation of sexually explicit deepfake images and I very much welcome that move.

The government, however, need to go further and ban the creation of all non-consensual deep fakes to protect us from political misinformation and help defend citizens from fraud online. Indeed, the Government should hold the whole supply chain responsible for deepfake creation and proliferation.

I am part of a group across the political spectrum which is lobbying the Government to take more action on deepfakes (Control AI) as I believe if we allow the creation and distribution of deepfakes to continue it will have an impact on democracy across the world, not least here in the UK.

It is tough enough being a politician in today's fast-changing world. We need a system that supports truth and factual content, not AI designed to do harm to those putting themselves forward for election.

Until then the message is, don’t always believe what you see and hear online – it may be true, but then again it may be a deepfake.