• Home
  • Opinion
  • AI Agents in the newsroom are assistants, not substitutes

AI Agents in the newsroom are assistants, not substitutes

Jacob Otieno, Editor-in-Chief at TUKO.co.ke.
Jacob Otieno, Editor-in-Chief at TUKO.co.ke.

Some time ago, one of the most consequential debates in the Artificial Intelligence (AI) community erupted over a question that is becoming increasingly difficult to ignore: Are we moving too fast with AI?

What made the debate particularly striking was not only the gravity of the issues being raised, but also who was raising the alarm.

Dario Amodei, the chief executive officer of Anthropic, called for the industry to slow down development of increasingly powerful AI models. His counterpart at OpenAI, Sam Altman, and Elon Musk, among others, endorsed the call. However, not everyone agreed.

Meta CEO Mark Zuckerberg proposed a more company-led approach to AI safety rather than a coordinated industry-wide slowdown. Nvidia boss, Jensen Huang, equally rejected the suggestion to slam the brakes on frontier AI development.

And then there was the chilling warning that preceded the debate. Jacob Coxon, a former Anthropic researcher, publicly warned that people building AI genuinely believe it could potentially kill humanity by the end of the decade.

These are extraordinary claims that we should neither casually dismiss nor simply accept as the gospel truth. However, despite the disagreement over how fast we should move, we can all agree on one thing: AI is here, and it is already changing and transforming the way we work.

AI adoption isn’t slowing down

As the tech giants debate the speed of development, another AI revolution is taking shape quietly in places far less glamorous than the AI laboratories. It is happening in our newsrooms.

AI is reshaping the way we do research, analyse information, create and package content, develop applications and software, and solve problems. It is driving productivity to heights that would have seemed unimaginable just a few years ago.

At TUKO.co.ke, we are not sitting around waiting to see what AI will do to journalism, or what it might become as we have been told. We are constantly experimenting with it. We are testing it. We are learning its limitations. And we are integrating it into our workflows and editorial processes.

Most recently, we introduced AI agents into our workflows. No, TUKO.co.ke is not replacing human editors with bots. Instead, it is equipping people to work alongside technology, a distinction that matters enormously amid the growing fear‑mongering over what AI may become.

Meet Leo, TUKO’s editorial assistant

Leo is not a real person. It is a name I’ve assigned to one of our AI agents at TUKO.co.ke for this engagement. Leo serves as an editorial assistant.

Leo can analyse news sources and information supplied in all formats, including documents, screenshots, audio and video. 

He can transcribe interviews, process large volumes of information, produce summaries for journalists and help turn raw information into an initial draft, perfectly aligned with our editorial guides. He handles some of the work that tends to consume enormous amounts of a journalist's time without necessarily being the part of journalism that creates the most value.

Leo does not get tired. He can work 24 hours a day if you want him to, handle multiple tasks at extraordinary speed, and he can assist several team members simultaneously without becoming overwhelmed by the workload.

But there is something Leo cannot do, and that is making editorial judgements. He does not decide what stories we should cover. He does not build relationships with sources. He does not understand the human consequences of a story in the way a good journalist would. And he certainly does not carry the editorial responsibility for what we publish. That responsibility remains squarely with our human editors.

A more productive newsroom

Since introducing Leo into our editorial workflow, we have observed a remarkable increase in efficiency and productivity. The number of stories we are able to publish in a day has increased by roughly 40%.

Take one of our journalists, whom I will call Natali, not her real name. Before we introduced the AI agent, Natali could produce about six articles in a day, or about 30 in a week. Today, she produces around 10 stories daily and close to 50 in a week.

Leo has helped speed up the editorial process that once consumed most of Natali’s time. On average, she now spends just 20 to 40 minutes moving a story from draft, writing, and packaging to publishing and distribution across our platforms. 

That level of efficiency gives Natali time to focus on research, investigations, interviews, following leads, and uncovering the story behind the story, which is what serious journalism truly demands. Just as importantly, Natali’s work-life balance has improved.

Back to the AI debate

The big conversation about AI seems to focus more on whether machines will take people's jobs. Perhaps we should also ask a different question: What if AI can help human workers do a better job while improving their work-life balance?

That is the opportunity we see at TUKO.co.ke. As with revolutionary and disruptive technologies that came before some of us were born, AI should not be perceived any differently. 

Lest we forget, Johannes Gutenberg’s printing press revolutionised publishing; the telephone transformed interpersonal communication; the internet reshaped news distribution and reporting; and the smartphone transformed how we consume information.

Those who refuse to embrace AI may eventually find themselves competing not against the bots and robots, but against those who know how to leverage the new technologies. Therefore, our response to AI cannot be pegged on fear. We have to adapt, quickly.

The non-negotiables

Adaptation does not mean blindly embracing every new AI model that comes our way. Safeguards must stay in place. Human oversight is a mandatory requirement. These are not my words. The industry regulators, like the Media Council of Kenya, insist on human oversight when using AI in the newsroom.

The warnings from people working inside the AI industry deserve serious attention. But responsible innovation does not mean staying away from innovation.

For those of us in the business of news, the AI agents are already here. They are not here to replace us. They are here to assist us.

Disclaimer: The writer is Jacob Otieno, Editor-in-Chief at TUKO.co.ke. He is passionate about championing journalists’ rights and media freedom, and also writes on the role of technology in shaping the future of news media. Views expressed here are his own.

Share



ITWeb proudly displays the “FAIR” stamp of the Press Council of South Africa, indicating our commitment to adhere to the Code of Ethics for Print and online media which prescribes that our reportage is truthful, accurate and fair. Should you wish to lodge a complaint about our news coverage, please lodge a complaint on the Press Council’s website, www.presscouncil.org.za or email the complaint to enquiries@ombudsman.org.za. Contact the Press Council on 011 484 3612.
Copyright @ 1996 - 2026 ITWeb Limited. All rights reserved.