I write high-speed, strategy-driven Military Sci-Fi & Space Opera, where cunning commanders, elite soldiers, and alien warlords fight for survival on the fringes of space. Subscribe to my newsletter for my latest updates!
Now You See Me
Now You See Me(5 minute read) Will We Know What Is Real?I have written before about artificial intelligence and where it might take us over the next few decades, but there is one part of the subject that deserves a newsletter of its own. How will we know what’s real? That sounds like a slightly philosophical question, but it is rapidly becoming a very practical one. We are already beginning to see deepfake videos that can put words into the mouths of politicians and celebrities. Artificial intelligence can reproduce a person's voice from surprisingly small samples of their speech. Photographs can be created showing events that never happened, involving people who were never there, in places they have never visited. Some of them are still fairly easy to spot. The hands are wrong. The lips do not quite match the words. Someone has six fingers. The lighting looks peculiar. A supposedly famous person sounds almost, but not quite, like themselves. Unfortunately, that’s probably temporary. AI-generated images from only a few years ago now look laughably crude compared with what can be produced today. The same thing is happening with video and voices. If that improvement continues for another ten years, never mind fifty, spotting a fake simply by looking at it may become almost impossible. That creates some obvious opportunities for criminals. Imagine receiving a video call from your daughter. She looks exactly right. She sounds exactly right. She tells you she has lost her phone while travelling and urgently needs $2,000 transferred to a particular account. Today, many of us would probably believe her. Now imagine that there is no daughter at the other end of the call. A criminal has copied her face from photographs and videos on social media, cloned her voice and created a convincing digital version of her in real time. The traditional advice to beware of strange emails and suspicious links starts to look rather inadequate. It could become even more serious at the political level. Imagine waking one morning to a video of the President of the United States announcing that America has been attacked and military action has begun. The video appears on social media, is reposted millions of times within minutes and looks completely genuine. Or perhaps a recording emerges of a prime minister privately admitting to corruption. A military commander appears to order his troops to stand down. A chief executive announces that his company has created a product that is highly toxic to human health. A central banker announces an emergency financial measure. All of them could be completely fake. The dangerous part is not simply that we might believe something that isn't true. The bigger problem is that eventually we might stop believing things that are true. That could be even worse. If every embarrassing recording can be dismissed as a deepfake and every genuine photograph can be described as AI-generated, proof itself begins to lose value. A politician caught saying something appalling could simply respond, “That wasn't me.” A criminal recorded on camera could say the same. We could reach the strange position where photographic and video evidence becomes less trusted than it was a hundred years earlier. Fortunately, I don't think the answer is that we simply give up and accept that nobody will ever know anything again. Technology will probably create the problem, but technology can also help solve it. Future cameras could digitally sign photographs and videos at the moment they are taken. The signature could confirm which device recorded them, when they were recorded and whether the file had subsequently been altered. News organizations might use authenticated cameras whose recordings can be independently verified. Governments could cryptographically sign official broadcasts. Individuals might have verified digital identities allowing other people to confirm that the voice, message or video call really came from them. The important change would be subtle but enormous. Today, when we see a photograph or video, our first instinct is usually to ask: “Does this look real?” By 2076, that question may be almost meaningless. A fabricated video may look absolutely perfect. The lighting will be right. The voice will be right. Every facial movement and reflection will be right. There may be nothing visible that gives the deception away. Instead, we may ask: “Can this prove that it’s real?” That could lead to a whole new system of digital trust. A genuine news video might carry an unforgeable chain showing who recorded it, which organization received it and whether anything was changed before publication. A photograph without that chain would not necessarily be fake, but we would know it was unverified. Perhaps our phones will automatically warn us: AUTHENTICATED SOURCE or ORIGIN CANNOT BE VERIFIED The same principle could apply to telephone calls. If someone claiming to be your bank calls you, your phone might independently verify that the call genuinely originated from the bank rather than merely displaying its telephone number. If your daughter calls asking for money, perhaps her identity will be cryptographically verified before you even answer. That sounds complicated, but so did internet banking, digital passports and facial recognition when they first appeared. Within a generation, children may find it extraordinary that we once trusted photographs simply because they looked convincing. There is another side to this as well. We humans have always been rather good at believing things we already want to believe. Fake information doesn’t have to be perfect if it confirms someone's existing opinions. People frequently share stories today without checking whether they are true because the story supports what they already believe. Better artificial intelligence will make that problem worse. A fake photograph showing a hated politician accepting a suitcase full of cash will spread far more quickly among people already convinced that the politician is corrupt. A fabricated video of an unpopular public figure insulting a particular group could travel around the world before anyone has time to establish whether it ever happened. The correction might reach ten million people. The fake might already have reached five hundred million. That is probably one of the biggest challenges we will face during the next few decades. Not artificial intelligence becoming evil. Artificial intelligence becoming all too convincing. I remain an optimist about most new technology. Human beings are extremely good at finding solutions once a problem becomes important enough. We created locks because people stole things, encryption because people intercepted messages and anti-fraud systems because people cheated banks. We will create ways of authenticating reality because eventually we’ll have no choice. The strange result may be that the world of 2076 contains vastly more artificial photographs, voices and videos than ours does today, while at the same time genuine information is actually more securely authenticated. We may trust our eyes less, but perhaps that will not be entirely a bad thing. For most of history, seeing was believing. In the future, seeing may simply be the beginning of the investigation. Because in a world where almost anything can be faked, reality could become one of the most valuable things we have. 'til next time, Brian BA Gillies Grab your copy of Unite the Galaxy TodayUnite the Galaxy, the next instalment in the Return to the Galaxy series, is now available to purchase on Amazon. Free on Kindle Unlimited. The story is heading toward the series finale, and the stakes have never been higher for Ewan, Shona, Velal and the rest of the crew.
***If you've been meaning to dive into the Return to the Galaxy Universe, now’s your chance!Whether you're a new reader or just haven’t grabbed your copies yet, now’s the time to catch up on the award-winning series readers are calling “better than Scalzi” and “the best book since Heinlein died.” If you haven’t started the series yet, Book 1 is still available to buy for 99 cents or read for free on Kindle Unlimited:
***Discover More Sci-Fi Adventures Each week, I’ll team up with a group of talented sci-fi authors to bring you an incredible selection of books. Whether you’re looking for thrilling space battles, deep space mysteries, or first-contact encounters, these books are packed with stories to fuel your imagination. Author Spotlight: |
BA Gillies
I write high-speed, strategy-driven Military Sci-Fi & Space Opera, where cunning commanders, elite soldiers, and alien warlords fight for survival on the fringes of space. Subscribe to my newsletter for my latest updates!
The Future of Entertainment (5 minute read) More Real Than Reality? I have written before about artificial intelligence, virtual reality and where both might take us over the next fifty years, but entertainment may be one of the areas where the change is most dramatic. Today we watch films, read books, play computer games and occasionally put on a virtual reality headset. By 2076, the distinction between watching a story and actually experiencing one may have become very blurred. Imagine...
Will We Still Love Our Cars? (5 minute read) For many years, I lived in Windsor and worked in central London, roughly twenty-five miles away. I normally drove to work. Because of the congestion, particularly once I reached central London, the journey could easily take an hour and a half each way. That meant I spent around 15 hours a week simply driving to and from work. At weekends, I probably used the car for another 3 hours. So, even though people regarded me as someone who used his car...
Will Humanity Live Beyond Earth by 2076? (5 minute read) When America celebrates its 300th birthday in 2076, will human beings be living on other worlds? Not merely visiting them for a few days, but living there permanently. Working, raising families, building industries and gradually creating communities that no longer depend entirely upon Earth. For most of my lifetime, that idea belonged firmly to science fiction. Astronauts reached the Moon more than fifty years ago, yet only twelve...