By Paul Cheung

Hacks/Hackers, a nonprofit organization fostering collaboration between technologists and journalists, hosted a series of “AI x Journalism” events at the recent SXSW tech conference in Austin, Texas. Exploring the potential impact of AI on the future of news and information, the discussions ranged from envisioning an AI-influenced media landscape in five years to the paradox of AI disclosure impacting audience trust, along with the ethics of generative AI and insights from over 40 journalism leaders at the forefront of AI.

The sessions also revealed anxiety among journalists about the potential of AI to replace human reporters. After all, the CEO of German media group Axel Springer did warn that journalists are at risk being replaced by AI if they don’t innovate. This fear coincides with the news industry’s persistent financial woes and shrinking workforce. In 2023 alone, the news media workforce lost over 20,000 jobs and 2024 is starting with a grim outlook.

Amid the conversations and debates around the opportunities and risks of utilizing AI in journalism, one thing that is becoming clear to me is that journalism is stuck in a cycle of contemplation about what to do and not do with AI. Meanwhile, other sectors like technology, banking, and consumer retail products are proactively preparing strategies, shifting resources, testing new products and creating new job functions for an AI-centric future.

The clock is now ticking for traditional journalism, with a critical window of 12 to 24 months to make significant strides in substantively incorporating AI.

Throughout my career at legacy news institutions, I observed a recurring pattern: a reluctance to fully leverage technologies or disruptors for transformation. A study by the Reuters Institute for the Study of Journalism in 2014 best captured this prevailing hesitance: “Journalists have an occupational ideology (a system of beliefs about what ‘real journalism’ is) with claims to an exclusive role and status in society, which keeps together their professional identity.” 

For instance, the rise of bloggers and social media influencers as significant players in audience engagement and news content creation was met with skepticism, despite their proven financial success and ability to engage large followings. Rather than seeking to understand and perhaps emulate these successes, there’s been a tendency to circle the wagons, debating the essence of journalism and gatekeeping the title of journalist.

This occupational ideology within journalism is also leading news organizations to adopt a sense of self-importance, further disconnecting the industry from the everyday experiences of their audiences. Look at the way the industry markets itself: slogans like “Democracy Dies in Darkness,” “Truth. It’s more important now than ever,” and “Journalism by Journalists. Not comedians.” These messages do not convey how journalism aims to inform and engage audiences by fulfilling their information needs. Instead, it can come across as preachy and detached from the realities many people face.

There’s also ongoing concerns about whether AI could deepen mistrust in news media and erode the relationship between audiences and reputable news sources. While tech companies definitely play a role in this trend, the journalism industry must also look inward, acknowledging its part in the decline of trust in news among audiences. The industry’s continued lack of commitment with diversity has left it disconnected from the communities it aims to serve. A 2023 Pew Study on Black Americans’ news experiences highlighted this disconnect. According to the report, just 14% of Black Americans are highly confident that Black people will be covered fairly in their lifetimes. This is not a technology problem but a lack of diversity problem in journalism.

This sense of self-importance and tendency to question rather than embrace new possibilities has led journalism to a state of inertia, preventing the industry from staying ahead of disruptive changes.

While discussions around AI ethics, audience trust, and copyright concerns are vital, we must be careful not to become so trapped in them that we neglect learning and iterating from experimentation. This is a crucial moment for journalism to directly engage audiences in testing new experiences, products, and pushing new boundaries of the craft in the era of AI.

It’s worth noting that the use of AI in journalism is nothing new. The Associated Press began using AI in 2015 to automate corporate earnings reports, while Richland Source, a local news outlet in Mansfield, Ohio, has been utilizing content automation technology since 2018 to enhance their sports coverage. Furthermore, recent initiatives by Microsoft with five journalism organizations and OpenAI with various news organizations, including AP, Le Monde, and Axel Springer, indicate a growing trend towards embracing these technologies for innovative journalism.

While most existing efforts around AI in journalism have focused on production issues like automated content generation or translation or creating derivative content for social media platforms, they have not grappled with the profound implications AI could have on journalism’s fundamental structure and operating models. Three key questions loom:

1. How will AI fundamentally transform the core responsibilities of journalists? For example, if AI can contextualize a journalist’s reporting and output customized story formats tailored to each audience member’s preferences, do reporters still need to construct full articles? In this scenario, what becomes the role of an editor when they are not editing a final, one-size-fits-all product?
2. How should news organizations prepare for the seismic cultural shifts, restructuring, and resource reallocations required in an AI-driven world of journalism? For example, what function will be tasked with overseeing AI’s ethical use in newsrooms? Will it be an individual role akin to a standards editor, or an entire team responsible for guiding principled decisions around AI applications – from story generation to audience interactions? Do news organizations also need to hire retrieval augmented generation (RAG) specialists to improve the efficacy of large language model applications by leveraging custom data? If so, who will oversee this effort – editorial newsroom leadership or business operations and what product is it supporting?
3. How can news organizations embrace an agile, iterative approach to developing new products and solutions, actively involving their audiences throughout the process to gain valuable feedback and insights? For example, what is stopping news organizations from inviting their audience to participate as test panelists for upcoming product launches, ranging from new newsletters to news apps or other niche products? This could help quickly identify potential flops and save you money and resources or uncover opportunities for pivots that can turn into successful revenue streams. Many innovative products today have been shaped through learning from mistakes and iterations. The business communication giant, Slack, was supposed to be a video game, Glitch, but when the game didn’t have enough users to support a business, the company pivoted.

This is nothing new for the news industry. Previous digital disruptions have created new job roles while phasing out other traditional journalism functions. I believe AI will have a similar impact. Instead of reflexively resorting to layoffs, what lessons learned can be applied by the industry from those past upheavals in order to thoughtfully realign resources, reskill existing personnel, and cultivate new areas of expertise.

Tackling these deeper questions is critical, as AI’s disruption extends far beyond just production efficiencies. This is the time news organizations must think critically about how to redesign their operations and workforce models to harness AI’s potential, while retaining journalism’s core editorial expertise and values. Complacency and superfluous policy debates risk leaving the industry unprepared for imminent transformation.

Fortunately, news organizations can leverage partnerships with organizations like CNTI, Online News Association, Hacks/Hackers, and others to collaboratively tackle the challenges of transforming their operations for an AI-driven future.

Paul Cheung is a Strategic Advisor at Hacks/Hackers and a Columnist for the Reynolds Journalism Institute