AI Copyright Discussion Needs Fresh Perspective, Suggests Prominent VC

Venture Capitalist Vinod Khosla argues that artificial intelligence (AI) copyright laws should prioritize output results rather than how it’s trained.

“To best think about the copyright issues in AI training, we should focus on output similarity and not training methodology,” the report stated.

Khosla Urges Focus on AI Output, Not Training Methods

In a comprehensive report, Khosla explains that materials throughout history have often been derived from earlier works:

“All humans train on cumulative learning from many past works by other humans. AI may train on just a larger set of past works and be subject to similar rules and constraints but no more and no different.”

Khosla suggests a different approach to the ongoing debate about legal suits against AI. Instead of how it’s trained, the focus should be on what it creates, especially if it’s similar to previous work.

“The same instructions multiple times to AI may result in very different outputs which disproves the notion of “copying”.

Creative industry participants have recently initiated legal actions against AI companies, questioning the use of their clients’ works in the training data.

On October 20, BeInCrypto reported that Universal Music Group (UMG) along with other publishers had launched a lawsuit against Anthropic, an artificial intelligence (AI) firm.

The dispute revolves around Anthropic’s AI model, Claude 2, allegedly distributing copyrighted lyrics. UMG contends that Anthropic is doing so without securing the necessary licenses.

The lawsuit argues that Claude 2 has generated phrases resembling existing lyrics, even without explicit prompts.

It additionally claims that the program possesses the capability to create lyrics similar to renowned artists. The filing cited various artists including Katy Perry and the Rolling Stones.

Meanwhile, technology firm Story Protocol has secured $54 million in funding to address copyright infringements caused by generative AI technologies.

The prominent VC firm Andreessen Horowitz led the funding round. Various other VC firms participated, which, in exchange for equity, are providing the funds to handle potential lawsuits.

Ongoing Talk about AI and Jobs

While artists primarily discuss copyright issues with AI, discussions also revolve around its potential job impact.

Recent data from Statista indicates that office and administrative jobs face the highest AI-related risk at 46%. The legal profession follows closely at 44%, with architecture and engineering coming next at 37%.

Industry employment at risk of automation by artificial intelligence (AI) in the United States in 2023, by industry. Source: Statista

On the other hand, there has been a notable increase in search results for AI jobs within the burgeoning technology sector.

Furthermore, on July 7, BeInCrypto reported that Google Trends data showed search interest in AI jobs has surged to four times the volume of crypto jobs at their peak points.

Source