News BuzzFeed’s AI-generated content experiment is a glimpse of a bleak future

Internet media company BuzzFeed, which makes short essays, quizzes, and lists like “18 Insanely Comfortable Clothes from Amazon That Got Us Through the January Blues,” is struggling. After going public in December 2021 at a valuation of $1.5 billion, the company’s stock price plummeted, losing 80% of its market value by the end of 2022. But on January 26, 2023, BuzzFeed CEO Jonah Peretti released a memo describing how the company will integrate OpenAI’s ChatGPT large-scale language model into its production pipeline. therefore, buzzingThe valuation of the stock has increased by 150%.

Founded as a nonprofit in 2015, OpenAI became a de facto Microsoft subsidiary in 2019 in exchange for access to the tech giant’s vast computing power. OpenAI is a leader in the form of artificial intelligence (AI) known as large language models (LLM). An LLM is a statistical model of language, created by machine learning, that incorporates the probability of occurrence of any given sequence of letters. LLMs can generate textual output in the form of probabilistic predictions based on input cues. Given the cue “It’s raining…”, the LLM might output “dark” or “windy”. A full-fledged LLM like ChatGPT can do a lot more, from generating qualified graduate thesis to working software code.

The market’s enthusiastic response to Peretti’s memo provides an example of the tech industry embracing buzzwords and going with the flow. But it also offers an opportunity to consider the capitalist context of AI production and use, and the hollowness of industry-driven discourse about AI’s social impact and ethical dimensions.

focus is on profit

Since around 2015, the US tech giants leading AI R&D—Alphabet (Google), Meta (Facebook), Microsoft, Amazon, and Apple—have made a concerted effort to portray their AI as “ethical” and provide paint a bright picture. A future in which artificial intelligence improves everyone’s lives. Academia has also stepped in, offering theories and models to make AI more ethical.

The “ethical AI” discourse tends to ignore the industrial context of AI. In the simplest terms, this means recognizing that artificial intelligence is largely a commodity. This means it is produced to be sold to generate a profit. Since not all applications of AI are profitable, this simple insight has major implications for thinking about AI, its social and ethical implications, and possible futures. It is for this reason that the industrial context of AI has been excluded from industrial and academic discussions in support of industry. Consequently, AI is often considered a discrete technology whose ethical and social implications can be dealt with on a purely technical level, certainly without the need for historical and systems analysis of capitalist economies.

However, the driving force of capitalist dynamism is not technology, but the competition between capitalist enterprises, which, as Karl Marx and Frederick Engels put it, drives them to “constantly innovate the means of production”. “. Automation, the use of technology to increase labor productivity, is also driven by competition. If your company can use ChatGPT to produce three high-impact articles with the same amount of labor time it takes my company to produce one article by hand, then your company will have a decisive competitive advantage.

what BuzzFeedWhat the memo tells us about AI

ChatGPT provides truly amazing capabilities for algorithmic manipulation of languages. It represents a major scientific and engineering triumph that could ultimately tell us a lot about how human cognition works and the differences and similarities between the way humans and machines process information. Peretti’s memo ignited investor enthusiasm, describing a mundane application that would move “artificial intelligence-inspired content… And personalize our content for our audience.” In short, “maximize the creativity of our writers, producers, and creators.” In other words, increase the productivity of labor engaged in clickbait.

A narrow focus on increasing productivity leads to contradictions and empty discourse, because the main goal of capitalist production—capturing value—must remain self-evident. Google AI’s homepage proclaims its commitment to artificial intelligence to be “socially beneficial, fair, responsible, and for all.” It doesn’t mention that details like this are only possible if Google can continue to generate revenue streams. In particular, it fails to mention that Google’s revenue stream (and that of other large tech companies), as many studies have shown, are based precisely on irresponsibility, a lack of concern for societal interests, and a very over vague explanations. And the aggressive acquisition of personal data – what Shoshana Zuboff calls “surveillance capitalism”.

This emptiness manifests itself in BuzzFeed The memo when Peretti described the company as aspiring to be “the premier platform for creatives to do their most inspiring work, reach the most people, and live their best lives,” and just to follow up, in the next paragraph, declares BuzzFeed Expand “from AI-driven curation (feed) to AI-driven authoring (content).” In other words, from using AI to recommend handwritten articles to users, BuzzFeed will use artificial intelligence produce those articles. Peretti goes on to describe a “new era of creativity, where creative people like us play a key role, delivering ideas, cultural currency, inspirational cues, IP and formats, all made possible using the latest technology. The idea isn’t to replace workers entirely — the company says it will lay off 12 percent of its workforce by 2022 — but to have them feed ChatGPT with something new so it can keep making clickbait. Here, the appeal of the creator economy — its authenticity — is as real as possible, No creator.

But that’s not the only contradiction at hand. A more obvious manifestation is in the “ethical artificial intelligence” discourse promoted by industry and adopted by academia and government. Google asserts that “value-based AI is good for your business.” One might reasonably ask: what values? Are they Confucian, Christian, socialist, liberal or something else? The answer, of course, is values ​​that do not conflict with the demands of capitalist production. Ethical AI is a dead end endeavor doomed to failure on its fundamental terms, as it refuses to acknowledge the fundamental drive of capitalist production to integrate AI into labor processes and simply assumes that more automation is better for everyone .

This BuzzFeed The memo makes clear that the stakes in introducing AI into business processes are economic, not moral. Peretti described the need for a “strong focus on delivering strong value to our partners so that they continue to spend with us through the recession.” He made it clear that improving labor productivity was a clear goal, noting that “in difficult economic times , we need to … save every penny of cost.” The dubious nature of the AI ​​industry’s moral claims doesn’t need to be acknowledged by the radical left. As early as 1970, the neoliberal economist Milton Friedman pointed out that the only “social responsibility” of business is to “increase profits.”

This BuzzFeed The memo might just be hype. AI strategic plans can fail. Or companies might rise to new profitable heights based on AI-generated clickbait. In either case, the memo paints a bleak picture of capitalism’s vision for AI: Trivial apps designed to wring more productivity scum out of labor, peddled alongside empty talk of social welfare and moral improvement.