I’m here to tell you you’re not alone – and I’m not just saying that because I’m really jet-lagged and writing this at 4am. All around the world, people are resisting the slide into authoritarianism, many through the simple act of reading or watching and sharing stories that uncover the truth.

That’s what I took away from the Global Investigative Journalism Conference I just attended in Kuala Lumpur, Malaysia. More than 1,000 investigative journalists came together for five days to share ideas on the big themes we’re all covering, as well as discuss how to do better work and bigger collaborations.

Autocratic systems are strengthening their hold around the world. There’s money controlling our politics and society, governments are backsliding on climate commitments and the rise of AI seems inexorable.

But is it? My favourite session at the conference was a talk with Karen Hao, the journalist whose book Empire of AI has helped break down what’s happening for those of us slightly overwhelmed by the explosion of AI.

Hao has also irritated AI company CEOs by essentially calling their projects a load of hot air – not something billionaires love to hear. Her central premise is that the AI goldrush is not about technological necessity. The AI advances we really need, like those in health or science, don’t rely on the crazed growth of large language models and an army of new data centres. LLMs and data centres are drawing on our water, land and energy at a time when those resources are getting scarcer and scarcer, potentially forcing us back towards dirty energy.

Hao argues that the AI boom is about empire building. The companies are striking deals worth tens of billions of dollars for extra computing power to build their control of key resources around the planet.

Here’s a quote from Hao’s talk that blew my mind:

Our own reporting on AI has already made some waves. The story on Jeffrey Epstein-styled chatbots being accessible to children prompted the company hosting them, Character.AI, to ban kids from the platform for good this week. Our exposé on an influencer peddling AI generated videos to stoke hate – and rake in cash – on Facebook has been picked up across the UK media. And this week, we’ve exposed a scam that’s using AI at every juncture – I’ll tell you more about that later on.

But before that, one more reflection on a week spent with the best and brightest of the industry: we can only expose the risks and harms of AI and the rise of authoritarianism if we can keep going.

Most successful newsrooms have a small but committed supporter base that keeps the journalism going week in week out. If you’re already supporting TBIJ financially, that’s you. If you can’t afford to, sharing our stories helps just as much.

And if you can, please do think about helping fund more investigations at this precarious moment. You can join our membership community, the Bureau Insiders, using the button below:

Factchecked!

Each week we reveal a fascinating fact from our reporting…

Did you know?

Last year, we won a lawsuit brought against us and our work.

Find out more

We beat a libel suit from a multibillion dollar company formerly linked to Kazakhstan’s ruling elite when the claim was dropped after two years.

We rigorously fact-check all our findings and will always defend our work. But a libel case in the London courts can easily cost upwards of £500,000 to defend – and well over a million if it reaches trial. For newsrooms like ours, the stakes are existential.

Read more here. 

AI fakers impersonate real lawyers online

Scammers on Fiverr are using AI tools to impersonate real solicitors in the UK – and dupe customers into handing over cash for chatbot-generated “legal advice”

Our latest investigation identified dozens of listings on Fiverr, one of the world’s biggest freelance platforms, offering legal services while impersonating genuine UK lawyers. (That’s a criminal offence, in case you were wondering.)

The names and professional registration IDs are real, but the headshots are AI-generated. Any legal advice given by the impersonators – some of whom seem to be using chatbots to generate their answers – is effectively worthless.

It’s another example of how AI can be used pretty easily by scammers. This isn’t just a problem for you and me, but also for the lawyers who are offering real services – including some who legitimately use Fiverr.

The question is, who’s responsible for making sure this doesn’t happen? If you work in this area and have thoughts, please get in touch. We’re always keen to learn more.

We reported our findings to Action Fraud and the Solicitors Regulation Authority (SRA), which then published a scam alert about Fiverr. Phil Brickell, a Labour MP who chairs the parliamentary group on anti-corruption and responsible tax, said that “the government’s forthcoming fraud strategy must include measures to hold online companies to account”.

We talked to victims at both ends of this scam: prospective clients and impersonated lawyers. In fact, one of the lawyers even decided to chat to his doppelganger, to find out just what kind of advice “he” was giving…