New EU digital regulations impose legal accountability for content
Ten months on since our first AI blog of the year and we have new legislation to abide by, courtesy of the EU whose Digital Services Act came into force a couple of months ago. As this summary in The Verge explains, this legal intervention on content accountability is only the beginning, leaving PR pros questioning – should we still be including it in our PR toolkit?
Generative AI as a content tool has always been under the editorial microscope at IBA, especially following the launch of ChatGPT in 2022 and we’re constantly on the lookout for how it is playing out. So our Social Media team has been keeping a close eye recently on TikTok following the introduction of the new EU Digital Services Act – where of course, AI is at the front and center of this development!
These new regulations imposed by the European Union are forcing TikTok and similar organizations to crack down on hate speech, illegal material and most importantly – disinformation. As explained by the New York Times, AI tools such as ChatGPT are today’s most powerful tools for spreading misinformation on the internet – including in the press and social media.
This news comes as no surprise to the team at IBA, who previously put ChatGPT under the test to produce content, which was then fact-checked by our in-house experts. With a number of false statements coming under the microscope – it’s clear that anyone making use of ChatGPT needs to proceed with caution, and a fine toothcomb before sharing any AI-generated content!
So, how is TikTok keeping up to speed with these new regulations you may ask? The platform is now prompting users to declare the provenance of their content – such as its legality and its source, is it AI-generated or not? The onus is back on the content provider. This all sounds a bit of a cross between smoke and mirrors, and passing the legal buck (whose ‘truth’? Generated by AI but vetted by humans? How do you define legal?), but it does reveal a tide-change in customer mindsets. Brands need to be aware that AI content generation may not exactly be a simple panacea to their content woes after all.
Remember what we’ve said before – PR 5.0 is all about the three eyes of AI – ideas generated by humans, written by humans, and read by humans!
It does mean that keeping organically up-to-date with the latest trends, and social media developments to ensure your content gains maximum attention and engagement is not enough. You need to stay on the right side of the law!
Staying on the right side of the AI legal battle is a smart move for TikTok – as new regulations, much like this EU Digital Services Act, are unfolding as I write, and will continue to do so.
For those still considering using ChatGPT and AI tools alike, here’s a timely reminder to keep it personal, emotive, and always proofread – it’s a sure-fire way to keep the AI police at bay!
Given enough monkeys with enough time and enough keyboards, the Infinite Monkey Theorem claimed that the monkeys could write the complete works of Shakespeare or at least Hamlet. Tech moved on from the monkeys.
The advent of ChatGPT has been inevitable. Technology was making plagiarism too easy. Google it, cut and paste it, edit it and woh, you have your own version of the truth. Enter ChatGPT that will do it all for you!
ChatGPT is the first of many
The media is all over ChatGPT – probably out of a guilty conscience. I’m going to leave aside the ethics and legitimacy of using such tools to cheat in examinations and coursework – cheating has always been done, technology just made it a little harder to detect. But the “cheats” that pass their finals magna cum laude will have to pass the life test!
My attention is on human generated copy and the growth of Marketing Speak into copy that already sounds as if a machine generated it.
Look at these statements, generated way back in 2014 by the original Business Buzz Word Generator pioneered by the Wall Street Journal.
“Next quarter we will launch our new alignment-killer which will silo the vector holistically” and “Moving forward the marketplace has changed. Gamify vertically or innovate.” Funnily enough, nearly 10 years on, and they still resonate!
Language must live
Now don’t get me wrong. I’m all for the development of language and the power of words to change attitudes as in female emancipation versus gender, celebrities versus influencers, and I look forward each year to the new words added to the Oxford and Merriam Webster dictionaries – reserving my right to not agree with all of them.
PR language has to evolve – the drumbeat, rolling thunder, tangible business outcomes, low hanging fruit, fleshing out detail, singing from the same hymn sheet, nuggets of information – love them all (apart from the “tangible business outcomes” which still bewilders me).
The problem isn’t that machines write like humans – it’s that humans often write like machines
ChatGPT is a wake-up call to stop writing copy in PR Speak, Marketing Speak, Corporate Speak, and to start using words to convey ideas.
There are a number of Marketing and PR Pros, and hacks, that truly do deserve to be quaking in their PR shoes, only too easily writing in corporate Marketing Speak. ChatGPT is the ultimate Wordsmith – and PR Agencies and Marketing Pros must be more than Wordsmiths – a demeaning descriptor for any writer. The implication is that writers, bloggers, content creators just produce copy that has been cranked out by someone with no interest in the subject being paid by the word. A machine can do that!
Writing comes with intention, and eyeballs is everything
But in PR and Marketing, we’re mostly writing about ideas and concepts, solutions and issues, and comments that shed light on issues.
Yes, we’re using words, just like any AI bot. Of course this requires coherence, something that ChatGPT excels at. But coherence is not enough.
In marketing we’re not trying to pass exams, we’re trying to influence, build marketing and corporate messages, deliver hidden agendas to our press releases that get across more than just a product launch. We’re trying to win the award, drive (often several overlaid) company messages – green adoption, ESG strategies, citizen data scientist, whatever, to pioneer change – and demonstrate corporate leadership in that change.
In marketing, coherence is not enough
Communication for marketeers is more complex and more precise. Often a single piece of press material needs to support a number of different company messages and at the same time to sell itself to several different precisely targeted audiences – an editor, a potential buyer, and a C-level ratifier.
Getting this right requires more than coherence. Tools such as robots and AI have always had a role, from the factory floor to the trading floor. The same applies to marketing. But communicating the success of our clients also needs the mixture of flare, discipline, and emotional intelligence that comes instinctively to top marketing professionals and to top salesmen.
Postscript
ChatGPT can be so persuasively wrong that Stack Overflow, a platform for developers to get help writing code, banned users from posting answers generated by the chatbot. “The primary problem,” wrote the moderators, “is that while the answers which ChatGPT produces have a high rate of being incorrect, they typically look like they might be good.” Exam sitters take note!
ChatGPT could also probably make a convincing fist of drafting the now essential apologies that seems to be required of politicians and celebrities.
Also, I understand that ChatGPT is not good at jokes so the growth in Slam Poets is safe, and apparently ChatGPT can’t meme – yet!
Judith Ingleton-Beer is CEO at IBA International
Leave a comment