For Christmas I received an interesting gift from a pal - my very own "very popular" book.
"Tech-Splaining for Dummies" (terrific title) bears my name and my picture on its cover, and it has glowing reviews.
Yet it was completely composed by AI, with a few easy triggers about me supplied by my good friend Janet.
It's an intriguing read, and really amusing in parts. But it also meanders quite a lot, and is somewhere in between a self-help book and a stream of anecdotes.
It simulates my chatty style of composing, however it's also a bit repeated, and very verbose. It might have exceeded Janet's triggers in collecting information about me.
Several sentences begin "as a leading innovation journalist ..." - cringe - which could have been scraped from an online bio.
There's also a mystical, repetitive hallucination in the type of my feline (I have no family pets). And there's a metaphor on almost every page - some more random than others.
There are dozens of companies online offering AI-book writing services. My book was from BookByAnyone.
When I got in touch with the president Adir Mashiach, based in Israel, he told me he had sold around 150,000 personalised books, generally in the US, because pivoting from putting together AI-generated travel guides in June 2024.
A paperback copy of your own 240-page long best-seller expenses ₤ 26. The firm uses its own AI tools to produce them, based upon an open source large language model.
I'm not asking you to purchase my book. Actually you can't - just Janet, who developed it, can order any additional copies.
There is currently no barrier to anyone creating one in anybody's name, consisting of stars - although Mr Mashiach states there are guardrails around abusive content. Each book contains a printed disclaimer stating that it is imaginary, created by AI, and developed "entirely to bring humour and happiness".
Legally, the copyright belongs to the company, but Mr Mashiach stresses that the product is intended as a "customised gag gift", sciencewiki.science and the books do not get offered further.
He wants to widen his range, creating different genres such as sci-fi, and maybe offering an autobiography service. It's developed to be a light-hearted form of customer AI - offering AI-generated goods to human consumers.
It's likewise a bit scary if, like me, you compose for a living. Not least because it probably took less than a minute to create, and it does, certainly in some parts, sound much like me.
Musicians, authors, artists and stars worldwide have actually expressed alarm about their work being utilized to train generative AI tools that then produce similar content based upon it.
"We should be clear, when we are talking about information here, we actually indicate human developers' life works," states Ed Newton Rex, founder of Fairly Trained, which campaigns for AI firms to respect creators' rights.
"This is books, this is posts, this is images. It's masterpieces. It's records ... The entire point of AI training is to find out how to do something and then do more like that."
In 2023 a tune including AI-generated voices of Canadian singers Drake and The Weeknd went viral on social media before being pulled from streaming platforms since it was not their work and they had not consented to it. It didn't stop the track's developer trying to nominate it for a Grammy award. And despite the fact that the artists were fake, it was still wildly popular.
"I do not believe the use of generative AI for innovative functions should be prohibited, but I do think that generative AI for these functions that is trained on people's work without authorization ought to be banned," Mr Newton Rex includes. "AI can be very powerful however let's build it ethically and relatively."
OpenAI states Chinese rivals using its work for their AI apps
DeepSeek: The Chinese AI app that has the world talking
China's DeepSeek AI shakes market and damages America's swagger
In the UK some organisations - including the BBC - have chosen to obstruct AI developers from trawling their online content for training purposes. Others have actually chosen to team up - the Financial Times has actually partnered with ChatGPT creator OpenAI for instance.
The UK federal government is considering an overhaul of the law that would permit AI designers to use creators' material on the web to assist develop their models, unless the rights holders decide out.
Ed Newton Rex explains this as "madness".
He points out that AI can make advances in areas like defence, health care and logistics without trawling the work of authors, reporters and artists.
"All of these things work without going and changing copyright law and ruining the incomes of the nation's creatives," he argues.
Baroness Kidron, setiathome.berkeley.edu a crossbench peer in your house of Lords, is also highly versus removing copyright law for AI.
"Creative markets are wealth creators, 2.4 million jobs and a whole lot of delight," states the Baroness, who is likewise a consultant to the Institute for Ethics in AI at Oxford University.
"The government is weakening among its finest performing industries on the unclear guarantee of growth."
A government spokesperson said: "No move will be made until we are absolutely positive we have a useful plan that provides each of our goals: increased control for best holders to assist them certify their material, access to top quality material to train leading AI models in the UK, and more transparency for right holders from AI developers."
Under the UK federal government's brand-new AI strategy, a national information library consisting of public data from a vast array of sources will likewise be made offered to AI researchers.
In the US the future of federal rules to control AI is now up in the air following President Trump's return to the presidency.
In 2023 Biden signed an executive order that intended to improve the safety of AI with, among other things, firms in the sector required to share details of the functions of their systems with the US government before they are released.
But this has actually now been repealed by Trump. It remains to be seen what Trump will do rather, but he is said to desire the AI sector to deal with less policy.
This comes as a variety of claims against AI companies, and particularly versus OpenAI, continue in the US. They have been taken out by everyone from the New york city Times to authors, music labels, and even a comedian.
They claim that the AI companies broke the law when they took their material from the web without their authorization, and used it to train their systems.
The AI business argue that their actions fall under "reasonable usage" and are therefore exempt. There are a number of aspects which can constitute fair use - it's not a straight-forward meaning. But the AI sector is under increasing scrutiny over how it collects training information and whether it ought to be paying for it.
If this wasn't all enough to contemplate, Chinese AI company DeepSeek has shaken the sector over the past week. It became the most downloaded complimentary app on Apple's US App Store.
DeepSeek declares that it developed its innovation for a portion of the rate of the similarity OpenAI. Its success has actually raised security concerns in the US, and threatens American's existing supremacy of the sector.
As for setiathome.berkeley.edu me and a profession as an author, I believe that at the moment, if I really want a "bestseller" I'll still have to compose it myself. If anything, Tech-Splaining for Dummies highlights the current weak point in AI tools for bigger tasks. It is complete of inaccuracies and hallucinations, and it can be rather tough to check out in parts due to the fact that it's so long-winded.
But offered how quickly the tech is progressing, I'm not exactly sure for how long I can stay confident that my considerably slower human writing and editing skills, are better.
Sign up for our Tech Decoded newsletter to follow the biggest advancements in global technology, with analysis from BBC correspondents all over the world.
Outside the UK? Register here.
1
How an AI-written Book Shows why the Tech 'Frightens' Creatives
Ada Koehler edited this page 2025-02-11 12:30:41 +08:00