First Publishing House Partnership: Axel Springer is the first global publishing house to partner with OpenAI for integrating journalism with AI technologies.
Enhancing Journalism with AI: The partnership aims to strengthen independent journalism in the AI era, enriching ChatGPT user experiences with recent and authoritative content.
Content Integration in ChatGPT: ChatGPT will include summaries of selected global news from Axel Springer’s media brands like POLITICO, BUSINESS INSIDER, BILD, and WELT, including paid content.
Transparency and Attribution: ChatGPT responses will attribute content to Axel Springer and provide links to full articles.
Supporting Axel Springer’s AI Ventures: The partnership also involves using Axel Springer’s content for advancing OpenAI’s large language models.
Goal of Partnership: Exploring AI’s potential in journalism to improve quality, societal relevance, and the business model of journalism.
OpenAI’s Commitment: OpenAI commits to collaborating with publishers and creators globally, ensuring they benefit from advanced AI technology and new revenue models.
About Axel Springer: A media and technology company active in over 40 countries, Axel Springer aims to be the global leader in digital content and classifieds. The company, based in Berlin, employs over 18,000 people worldwide.
Note: this topic originally linked to the Reuters article but I updated it with the official blog post later.
As part of the deal, when users ask ChatGPT a question, the chatbot will deliver summaries of relevant news stories from Axel Springer brands including Politico, Business Insider, Bild and Welt. Those summaries will include material from stories that would otherwise require subscriptions to read. The summaries will cite the Axel Springer publication as the source, and also provide a link to the full article it summarizes.
The content will get a “favorable position” in ChatGPT search results, with the goal of helping to drive traffic and subscription revenue to Axel Springer brands, according to a source familiar with the deal.
Selective/bias sources… Not sure how I feel about this.
Regarding the “who” I don’t think it can matter at this point in time since they only have one partner from the news publishing industry. One would need a fair representation of the whole industry to let every user choose their own bias as desired.
But maybe this is a important signal to the other players to enter the field. For all we know ChatGPT is a app with 100 million users per month and as such a potentially major distribution channel.
And looking at the development from this perspective one could argue that custom GPTs are something like the modern version of the website and we are in the early stages of the internet.
If OpenAI can drive more users to their app and shares the revenue with the builders this could potentially become a huge first deal.
But time will tell where this is heading. Today I want to see the actual implementation and find my personal use cases for this feature.
When I ask for information from ChatGPT via web browsing I expect it to go through the records without any sort of third-party influence. As a default state, ChatGPT will use their articles. This by definition is bias.
bias: prejudice in favor of or against one thing, person, or group compared with another, usually in a way considered to be unfair
If I decide to say to ChatGPT “Search this only using New York Times”, that it my conscious decision to be bias. Similar to Google.
This seems to be exactly what OpenAI is intending. Either join up, or get left in the dust. Tech-bros are so obsessed with disruption that they have no concept of stability.
Which is incredibly scary. If OpenAI controls custom GPTs - which turn out to be the “homepage of the world”, and ChatGPT has bia$ in its authoritative, consolidated, paraphrased answers, we are truly in for dark times.
Not at all. Journalism will be reduced to fact-collecting while AI will write it and distribute it.
Journalism died after Watergate. A prof told me after that event a LOT of people who just wanted to be famous enrolled in J-school… we’ve been on a downhill trajectory since… The problems go back further, though, if you dig… But yeah, I agree that journalism will likely suffer, but the companies will stay “relevant” a bit longer…
The data argument was the initial spark that got me thinking in this direction.
It could be that this is one of the previously announced data partnerships and Axel Springer’s management somehow turned this into a publishing partnership.
But with or without ChatGPT these outlets have had their fair share of issues and I surely hope there will be more competition for which news will be used for individual readers.
That’s what I want to see if we can’t access a version of the model with less alignment to values that I don’t know.
I am fully for OpenAI working with Journalists and Media companies. I would have no issue if, for example they had a custom GPT that appeared at the top as “Sponsored”. Like, “Business Insider GPT, news etc GPT”. Similar to Google Ads.
What I am against is ChatGPT / web browsing having inherent, non-disclosed bias in the media that it chooses from because of partnerships
I want to see the implementation first. We all did have our fair share of ups and downs with OpenAI. It will be interesting to see where this is heading…
And regarding the big tech companies. This is a major issue, in the sense that it is a dilemma. As of today we know that these AI models pretty much gobbled up all available books and news articles and what ever has been published without actually going ahead and buying a single copy of these inputs. And additionally, recently judges and courts seem to consent that this is perfectly fine.
What I want to say is: the train has left the station and we are on-board.