News media companies must take a stand against AI

This article originally appeared on the International News Media Association’s Ideas Blog.

In his Declaration of the Independence of Cyberspace in 1996, American poet John Perry Barlow wrote: “Governments of the Industrial World, you weary giants of flesh and steel, I come from Cyberspace. On behalf of the future, I ask you of the past to leave us alone. You are not welcome among us. You have no sovereignty where we gather.”

He largely got his wish. Governments worldwide enacted legislation exempting Internet platforms from laws that otherwise apply in the world of flesh and steel he despised.

We have been living with the consequences ever since. Digital platforms have operated on largely self-written, self-serving rules, creating an environment optimised for their profits. Unencumbered by many laws that still bind the rest of us, they have thrived.

The information superhighway we were promised in the 1990s is here, albeit riddled with potholes, diversions, dead-ends, and bandits happily and profitably peddling disinformation, clickbait, deep fakes, malware, spyware, ransomware, and scams. We have become used to finding what we’re looking for buried under promoted links or mounds of ads.

We all need to keep our guard up, all the time, online. Barlow’s digital utopia has become a digital swamp.

The lesson of the Internet to date is this: If you abolish rules, you breed anarchy.

What it means for the news industry

The news media industry has struggled to survive. The big platforms leached away its advertising income while other credible revenue streams remain frustratingly elusive. New competitors for peoples’ attention have commoditised its content, even as audiences have become larger than ever.

Despite remaining an essential part of any functioning democracy, the news industry has been relentlessly diminished by the Internet.

Fears are growing that this pain will become even more acute with generative AI. This technology’s automated output is indistinguishable from that of creative people. It exploits the work, investment, and creativity of human creators without asking — and without paying. This is a truly parasitic technology.

Should AI firms be allowed to leach away creative content and revenues, converting them into income as the big platforms have done with advertising?

Of course not. The implications are huge and, unlike during the formative phase of the Internet, they are also obvious.

This is why the creative industries have moved quickly to prevent history repeating itself. Musicians, photographers, artists, filmmakers, authors, publishers, performers, actors … every sector whose existence depends on earning money from creativity has voiced outrage at the industrial theft that became obvious as generative AI tools were released into the wild. Many moved quickly to protect their work. The scene is set for a series of showdowns, in courts and legislatures around the world, to try to define the rules for this new technology.

The power of copyright

The central issue, as I see it, is simple. While scraping content (copying it from the Internet and stashing it in private databases) is commonplace, that doesn’t make it legal. AI is doing the same thing as search engines — making copies of what it finds online — but what AI does with the content isn’t the same. What it is doing is certainly not ethical or, in my view, legal.

I believe that, when it comes to AI, copyright will prevail. AI training, after an anarchic start, will have to be done within the law; permission will need to be sought and given. A framework, technology, and marketplaces to enable this will emerge.

There are good reasons to be optimistic about this point. Even if laws vary from one territory to another, any AI business that wants to operate across them must comply with the most stringent to avoid legal peril. Additionally, AI firms will need to reassure commercial customers that using their systems does not create legal risks for customers. This, in turn, will need the certainty of licences and contracts to underwrite their warranties.

This promises that the next generation of AI systems will be better, fairer, more transparent, more competitive, and, most importantly, legal. AI firms will become more diverse, specialised in servicing different verticals in different global markets and performing different functions, and therefore becoming more competitive.

With control of the rights that copyright law gives them and that early Internet law substantially took away, creators will be able to choose which — if any — AI may use their work. They, and all of us, will be literally and figuratively richer.

Cleaning up the Web

But there is work to do. AI technology isn’t going to start behaving itself just because new rules get written (or old ones get re-asserted). Nor can we be sure when those rules are going to become clear.

That’s because AI technology isn’t limited to compliant companies in democratic jurisdictions. It’s also available to, and already being used by, those who care little about regulations emanating from the United States, Europe, and the UK and whose location keeps them out of reach. Those AIs are going to operate according to their own rules regardless.

Everything on the open Internet will be fair game to them. They’ll carry on using it however they want. On the Internet, as we all know, nobody knows you’re a dog — or a bot farm, or a government information warfare operation, or a criminal enterprise.

This means the use of AI for nefarious purposes is unstoppable. The creation of garbage will be supercharged by AI, so we can expect even more of it to flood online for the foreseeable future.

The Web, already dirtied and polluted, is going to continue growing murkier. Is it a lost cause? Not if we actively choose to change it.

Opportunities for change

One opportunity for change lies in creating online spaces that are not polluted. In these spaces, any AI use will need to be constructive and transparent, overseen by accountable humans. We can help people distinguish what is safe and trustworthy, and avoid unwittingly stumbling into the online mire.

Certainly, as the estate responsible for discovering, validating, and communicating new facts, the news media becomes indispensable — and the opportunity for it to get even stronger, and play an even more central role in democracy is huge.

First, we need to win on copyright. We need to start preparing for an era built around the rule of law, where the idea of permission exists, managing permissions is possible, and markets emerge on terms lawmakers can control.

Next
Next

Reflections on the INMA World Congress 2023