Microsoft’s AI chief made a bold statement, asserting that any content on the Internet is fair game for AI. However, this claim contradicts that most of Microsoft’s products, like Windows and OpenAI, are also available on the Internet in some form, protected by lengthy and complex terms and conditions. It’s unlikely that anyone at Microsoft would endorse the idea of their products being used for free just because they are online.
It’s important to note that just because something is freely available online doesn’t mean it’s free to use without any restrictions. For instance, licenses like the GPL, as Microsoft’s former CEO Steve Balmer pointed out, may allow you to use software for free but also require you to reciprocate. As Balmer described it, this interchange can be seen as a ‘cancer’ that ensures everyone’s freedom, highlighting the complex nature of digital ownership.
Therefore, it’s crucial to consider the potential risks of the proposed shift in digital ownership for organizations and their intellectual property. How will risk management and boards respond to these potential challenges?
Legal Landscape
As of this writing, the legal landscape around AI training data is complex and evolving. Microsoft, GitHub, and OpenAI are facing multiple lawsuits related to AI training data. The accusations range from a breach of contract to copyright violations, underscoring the intricate nature of this issue. The plaintiffs range from individual open-source developers to the New York Times.
Given the speed and complexity of the issues, general councils must monitor the situation and provide timely information to management and boards. Otherwise, judging the risk correctly and adapting strategies to mitigate the impact becomes impossible.
Social Contract of Free Content
A second impact is the social contract around content. Apart from the terms and conditions, our implicit agreement with sharing data, watching ads, or boosting credibility in return for content governs today’s Internet. Unfortunately, AI doesn’t watch ads, share valuable data, or influence our credibility. Thus, it doesn’t give the creator anything in return for the content it consumes.
Google’s new AI search feature is a particular problem in this area. They regurgitate the content found on the source website without compensating the companies or individuals who created the information. Thus, it destroys the livelihood of researchers, professionals, and journalists to the advantage of Alphabet’s bottom line.
There are two things to consider when evaluating marketing content and remuneration strategies. If you make money from the content, it would be whether you can afford to put it behind a paywall. Many AI companies are unwilling to pay for most of the content or, at most, pay the content aggregators such as Reddit or Stack Overflow.
Alternative distribution channels, such as social and branding, become critical for companies that use content for reputation-building and general marketing. You might win the game if the AI spits out your brand name, such as Kleenex, as a stand-in for paper tissue. Likewise, if most of the readers come from your channels instead of from AI, the company could insulate itself from the AI fallout.
The End of Free
Depending on the final verdict, we might see different outcomes in the courts and society. The courts may need permission from the copyright owners to use data for AI training. Congress might make a similar determination.
It is also possible that consumers will choose against AI offerings and vote in favor of traditionally created content. In particular, if faulty AI devalues the user experience in services like Google Search, the backlash might be too significant for AI initiatives to survive.
In all cases, we will see a change in how content is presented and monetized on the Internet. Boards outside the News business will need to be aware of the changing trends and their impact on any content an organization shares online. Sales and marketing pipelines might need to be adjusted to place whitepapers behind registrations and reduce the amount of material freely available for download.