Humanoid robot on air

"Technology alone is not enough. It’s technology married with the liberal arts, married with the humanities, that yields us the results that make our hearts sing."

In the short time it’s taken you to read that Steve Jobs quote, an AI ‘journalist’ agent has already finished writing an 800-word article on AI in journalism.

And by the time you finish this sentence, that same ‘journalist’ has written a 1200-word deep dive, referencing multiple sources.

Cause for cheer or concern?

As an agency, Four has embraced AI. At the core of this is a robust ethics-led AI usage policy and proprietary blueprint called The Difference Engine. We use AI on a daily basis for a range of purposes, from research to ideation, writing and most recently to support our media training. So, it makes natural sense that our counterparts in journalism are doing the same.

According to a recent Guardian report, AI is playing a part in some British journalists producing more than 100 bylines per day. And titan publisher Dow Jones is open about its use of AI and machine learning in complex data-driven investigations.

With newsrooms shrinking at pace, it’s not beyond the realms of possibility that that in future a news team might comprise 1-2 human staff, supported by a team of AI agents. Agents that can crunch big data in seconds. Agents that can interview people and file copy in a fraction of the time it takes us. The role of the modern journalist may shift towards that of programmer – but the key is that humans will still be involved.

The National Union of Journalists has, naturally, come out with a clear position: AI generated journalism cannot replace the talent, creativity or skill sets held by journalists.  And it is right. AI can struggle with complex contextual understanding, nuance and ethical considerations.

And while AI-supported journalism can bring many benefits in terms of efficiency, we must recognise the role of AI in the spread of misinformation and disinformation. AI tools make it easy for anyone to create fake news that can be hard to distinguish from accurate information, while even some trusted publishers have scored own goals through their use of AI to quickly create stories.

So, what are we learning?

The human touch: While AI can automate many tasks, the human aspect of media relations remains vital, particularly for specialist subjects. While relationships alone won’t cure a poor story, time spent socialising with key media will ensure your emails get read and give you an opportunity to bounce ideas around. This allows you to effectively test pitches and curate stories together with media. A win-win.

The pitch: Despite the above advice, it simply isn’t possible to know everyone. AI can mine deeply to understand a journalist’s interests, stance on issues and trends over time. When combined with the human touch, story pitches can be more tailored and personalised.

Transparency and ethics: We must ensure that AI tools are used responsibly and ethically. Be transparent about AI-generated content and avoid over-reliance on AI for tasks that clearly require human judgment. Do not let ‘efficiency’ become a byword for ‘lazy.’

Analysis: Modern AI sentiment analysis tools can achieve up to 90% accuracy. They are getting better at understanding context, detecting sarcasm and recognising subtle emotional cues, which were challenging for earlier models. This is significant when dealing with sensitive projects and higher volumes of earned coverage where up-to-the-minute analysis is required.

A shift to owned: As communications professionals, earned media has traditionally been king, or queen. But considering a shifting (shrinking) media landscape, we are seeing a shift by smart communicators and their clients who understand the value of complementary earned and owned strategies. Dull company newsletters and magazines were once a tick box exercise read mostly by the company’s marketers but in today’s shifting landscape, this has changed. With a focus on the right channels, publisher-level rigour and quality combined with meaningful insights, owned is becoming increasingly powerful. Knight Frank's The View is a great example of this.

You can read more on the shift to owned in my colleague Paul Dalton-Borge's blog here.

Communications professionals and journalists alike will continue to harness AI's capabilities to enhance their work. The key is striking the right balance, ensuring ethical practices, and never losing sight of the irreplaceable human touch. As we navigate this evolving landscape, remember that technology is a tool, not a replacement for the creativity and integrity that define our professions.

Geoff Robjent, director, property