AI and Local Journalism: Navigating a New Frontier

Introduction

The media industry is in trouble – but what else is new? I have worked on journalism-adjacent issues for only five years, and this refrain has followed me throughout my rather short professional and academic career. The digital transition, global trust crisis, and polarization, democratic backsliding, the COVID-19 pandemic, to name but a few relatively recent "troubles", have all affected the sustainability of independent journalism in different, yet interconnected ways. Exchanging views with people who have centuries of collective experience in this sphere at the Milton Wolf Seminar on Media and Diplomacy in April 2024 has only reaffirmed this rather troubling outlook – every year the media seem to be facing greater and more complex problems.

Taking a broader view, though, the seminar’s talks and one-on-one conversations have underlined once again the vital importance of quality journalism in a time of information and world disorder. With disinformation named the "top global risk", and a record-breaking number of people going to the polls across the globe in 2024, ensuring that audiences everywhere have access to reliable, factual information is paramount – and the media, both old and new, are best positioned to play this important role.

Among journalism’s many current and past plights, one issue has drawn especially great scrutiny in Vienna (and, frankly, at all the other events and conferences I have attended over the past two years) – what happens to the ever-suffering media industry in the age of the much discussed but still confusing and even a little frightening “Generative AI”?

The AI Takeover

While many discussions in the media sphere over the past several years have focused on the positive applications of Artificial Intelligence - leveraging the technology to gain a competitive edge and streamline the work of newsrooms across all stages of media production and distribution - more recently, they have taken on a sinister edge. The rapid deployment of Gen AI, since the unveiling of Chat GPT in November 2022, is coming to be seen as a major threat to quality journalism and media sustainability.

One of the main risks of Gen AI discussed at the Seminar is its potential to take over the job of journalists. With increasing adoption of "synthesizers" like Chat GPT, Microsoft's Copilot, or Google's 'AI Overviews' search functionality, which produce neatly wrapped answers to a user's query from their vast databases, built by crawling the Web, there will be less and less incentive for people to seek out the original sources – media outlets included. Of course, these technologies are still imperfect – Google, for example, is suggesting that users add glue to pizza and citing satirical articles by The Onion as health advice – but the idea is, that as AI grows and develops, it will eventually be able to supplant the media.

As with everything concerning the “AI hype”, we should probably approach any maximalist claims like the prophesized “media extinction event” with a healthy degree of skepticism. While it is true that media is going through yet another transition (or has it been one single never-ending one?), accompanied by layoffs and a search for new formats and meanings, industry titans and large publishing conglomerates are on occasion able to stand their ground and even work with Big Tech. The New York Times, for example, is suing OpenAI over the scraping of their articles to train the large language model (LLM) that powers Chat GPT, suggesting the developers potentially infringed on NYT’s copyright. On the other hand, large media conglomerates Axel Springer and NewsCorp, among others, have recently struck deals with AI developers over the use of their content in training and user interfaces.

What I find more concerning is the potential effect of AI, and the deals between large publishers and developers, on media concentration and the livelihoods of local outlets, who are very often left out of this equation until the very last minute.

Local Media Woes in the Age of AI

At any given point in recent history, local media have been closer than average to a catastrophic “extinction event”. In fact, it has already occurred in some places – media scholars in the US, and now increasingly in Europe, have highlighted the more and more common “local news deserts”, where independent local outlets no longer exist. Where local news is still available, it is still “endangered” – the diminishing advertising revenues, difficulties with attracting and retaining talent in the face of increasingly tougher “global” competition, and lack of resources to keep up with technological change have hit the less affluent local outlets especially hard.

In the context of the looming challenge of AI, local media is at a double disadvantage. On the one hand, they lag behind their larger, better-resourced competitors in implementing AI-powered tools, which only serves to further the divide between them. On the other, the deals between big media conglomerates and tech companies are exclusionary by nature, and local media do not have the bargaining power to strike agreements of their own.

But perhaps not all is lost for local publishers. After all, the issue of Gen AI is an extension of past debates between the media and Big Tech companies over compensation for the use of journalists’ work. What lessons can we draw from the past forays into compensation for news from Big Tech?

News Bargaining: Depression or Acceptance?

The debates on the "fair price of news" have been ongoing essentially since the Internet's coming of age. Legacy media used to rule the advertising market, but in came intermediaries, who undercut their business model by capturing ad revenues. While these intermediaries – social media platforms and search engines – provided exposure to vast global audiences, journalism's profits lagged behind its growing reach. The media felt they were getting the raw end of the deal, and argued that they generate profits for Big Tech by providing users with the content they seek, and thus should get a share of the revenues. According to some estimates, platforms owe US publishers alone $14 billion a year.

These figures are, of course, hotly disputed by the technology companies. Moreover, the platforms counter the media's argument by claiming that the exposure and clickthroughs they provide to media websites are actually a valuable service to journalists. Following this logic, social media companies shouldn't be compensating media outlets, and in fact, are the ones who should be getting paid.

Similar to the current challenge of Gen AI, bilateral deals and litigation were entertained by some outlets and publishers. However, the growing recognition of the precarious status of the media, coupled with the political pressures to deal with the information disorder have given rise to more centralized government responses. In the past 3 years, regulators across the world have put forward and even enacted laws mandating platforms to negotiate a fair price with media publishers for hosting their content.

The Australian News Bargaining Code – the first of its kind, adopted in 2021 – has been relatively successful, redistributing around AUD 200 million among media outlets, including some smaller media. It comes with caveats, though. While the Code did generate some profits for Australian journalism across the board, most of the value was captured by dominant national media publishers, leaving local outlets at a disadvantage.

This first foray into news bargaining has provided a blueprint for others like it, who can build and improve upon it. For example, Canada’s Online News Act, passed in 2023, includes transparency obligations – the platforms and outlets will have to report on the deals struck, preventing excessive value capture by big media.

While the success of the Canadian approach remains to be seen, this underlines once again that regulatory intervention can empower local media in this debate – both vis-à-vis the platforms and other media outlets.

The Future of Local Media and AI Regulation

AI in general, and the interactions between AI in media in particular, are not yet addressed sufficiently at the regulatory level. Legislators should get ahead of the problem before it is enshrined in tech companies' offerings and seriously hurts the media who have not been able to negotiate with tech platforms bilaterally – small local outlets, that underpin media pluralism and serve an important purpose in the information ecosystem.

The recent regulatory forays into news bargaining provide both a blueprint and a cautionary tale. When considering how to properly compensate media and protect them from the impacts of Gen AI, lawmakers have to ensure that local media are sufficiently represented and treated fairly both by tech firms and their larger colleagues.

References

World Economic Forum, ‘The Global Risks Report 2024’, Insight Report (Geneva, 2024), https://www3.weforum.org/docs/WEF_The_Global_Risks_Report_2024.pdf.

Felix M. Simon, ‘Artificial Intelligence in the News: How AI Retools, Rationalizes, and Reshapes Journalism and the Public Arena’, Columbia Journalism Review, 6 February 2024, https://www.cjr.org/tow_center_reports/artificial-intelligence-in-the-news.php/.

Kylie Robison, ‘Google Promised a Better Search Experience — Now It’s Telling Us to Put Glue on Our Pizza’, The Verge, 23 May 2024, https://www.theverge.com/2024/5/23/24162896/google-ai-overview-hallucinations-glue-in-pizza.

Casey Newton and @crumbler, ‘Casey Newton (@crumbler) in Threads’, Threads (blog), 23 May 2024, https://www.threads.net/@crumbler/post/C7VGpYSPOgT.

Clare Malone, ‘Is the Media Prepared for an Extinction-Level Event?’, The New Yorker, 10 February 2024, https://www.newyorker.com/news/the-weekend-essay/is-the-media-prepared-for-an-extinction-level-event.

Michael M. Grynbaum and Ryan Mac, ‘The Times Sues OpenAI and Microsoft Over A.I. Use of Copyrighted Work’, The New York Times, 27 December 2023, sec. Business, https://www.nytimes.com/2023/12/27/business/media/new-york-times-open-ai-microsoft-lawsuit.html.

Gerrit De Vynck, ‘News Corp. Signs Deal with OpenAI to Show News in ChatGPT’, Washington Post, 23 May 2024, https://www.washingtonpost.com/technology/2024/05/22/news-corp-openai-ai-murdoch-altman/.

Sofia Verza et al., ‘Uncovering News Deserts in Europe: Risks and Opportunities for Local and Community Media in the EU.’ (LU: European University Institute, Centre for Media Pluralism and Media Freedom, 2024), https://data.europa.eu/doi/10.2870/741398.

Agnes Gulyas and David Baines, ‘Introduction: Demarcating the Field of Local Media and Journalism’, in The Routledge Companion to Local Media and Journalism (Routledge, 2020).

Anya Schiffrin and Haaris Mateen, ‘Google and Meta Owe US Publishers $14 Billion a Year’, 2023, TechPolicy. Press edition, https://www.techpolicy.press/google-and-meta-owe-us-publishers-14-billion-a-year/.

Fiona Martin, ‘Do Bargaining Codes Have a Future?’, Public Media Alliance (blog), 8 April 2024, https://www.publicmediaalliance.org/do-bargaining-codes-have-a-future/; Tama Leaver, ‘Going Dark: How Google and Facebook Fought the Australian News Media and Digital Platforms Mandatory Bargaining Code’, M/C Journal 24, no. 2 (28 April 2021), https://doi.org/10.5204/mcj.2774.

Martin, ‘Do Bargaining Codes Have a Future?’; Platforms and the Press: What Is a Fair Price for News?, International Journalism Festival (Perugia, 2024), https://www.journalismfestival.com/programme/2024/platforms-and-the-press-what-is-a-fair-price-for-news.

Peter Zimonjic, ‘What We Know so Far about Google’s $100M Media Fund’, CBC News, 1 December 2023, https://www.cbc.ca/news/politics/google-money-media-compensation-details-1.7046677.