The future of journalism with AI

There is a personal reason why discussing this book makes me feel somewhat awkward. It is a book about artificial intelligence, but also about people like me: individuals who work with words, who search for and verify information, and turn it into a headline or an article. Ever since generative AI arrived, they have been asking themselves: all well and good, but what do we do now?

“The Great AI Deception”, by Martina Pennisi, has given me a few more reasons not to lose heart about my future. However, it does not downplay the revolution currently underway. The book contains something that should give those in my line of work cause for concern: machines that write, translate, summarise, search and correct. There is also a chapter on the future of news. The easy answer would be to think that artificial intelligence will never be good enough. But that would be nonsense. It will get better. That is one of the book’s most reassuring messages. Pennisi forces us to reframe the question. Not ‘what will AI do?’, but ‘what will we still need to be able to do?’ Understanding that an answer is not the same as comprehension; knowing where information comes from; seeking out the original source; realising what is missing; formulating a better question; not confusing plausibility with truth; not delegating judgement. Certain skills we used to regard as secondary are now becoming central.

A machine can summarise a hundred articles in a matter of seconds. But someone must have written those articles, gone to a location, phoned a source, obtained a document, realised that a statement didn’t add up, and decided that a story deserved the front page. If journalism were to stop producing original information, the machines would also lack part of their raw material.

The work best suited to surviving artificial intelligence is not necessarily that which a machine cannot do, but that in which we learn to use it without becoming dependent on it. Marco Trombetti sums it up as follows: the difference lies in understanding whether you are working for the machine or whether the machine is working for you. The key word is responsibility. A professional finds fulfilment in their work if they retain control of the process, know what to delegate and are able to assess the outcome.

AI does not render the profession obsolete: it highlights what was truly important and what was merely drudgery. I would not feel nostalgic for the hours spent transcribing an interview, looking up a date or organising notes. If a machine can do it in thirty seconds, long live the machine. The problem arises when we think that eliminating the drudgery also means eliminating the discernment.

“The Great AI Deception” is therefore a critical book that ultimately has a strangely optimistic effect: it is not about defending humans from machines, but about understanding what still makes them necessary. The future of my profession lies here. Not in writing faster than a machine: that battle is lost before it even begins. But in seeing what the machine does not see, seeking out what is not yet in its databases, choosing what matters, and taking responsibility for a doubt, a headline, a mistake. Using artificial intelligence to strip away everything mechanical from our work and discover how much human labour remains. It is not a small amount. Indeed, reading this book, I began to suspect that it is the best part.