US Copyright Office weighs in (sort of) on AI

January 31, 2025

Good morning. OpenAI has become quite good at keeping itself in the news cycle with a seemingly constant stream of announcements and rumors, plus a regular dose of hype.

Well, they’ve managed it again. In talks to raise $40 billion, securing bigger government partnerships … there are, as always, plenty of unknowns. But we’ve got the details.

— Ian Krietzberg, Editor-in-Chief, The Deep View

In today’s newsletter:

💰 OpenAI in talks to raise $40 billion in funding at $340 billion valuation
🔭 OpenAI partners with U.S. National Laboratories; offers o1 for nuclear security
🏛️US Copyright Office weighs in (sort of) on AI

Source: Tines

AI adoption in the enterprise can feel like an uphill battle, as conflicting priorities, unclear ROI, evolving regulations, and security risks add layers of complexity. And in a sea of AI content, how do you separate the hype from real guidance?

That’s why Tines wrote Securing AI in the Enterprise: A step-by-step guide for IT and security leaders. This guide is designed to help IT and security leaders take a proactive, security-first approach to AI adoption, with practical strategies to address some of the biggest challenges.

What’s inside:

Insights from IT and security leaders at pioneering companies
Actionable steps for secure AI adoption that prioritize privacy and compliance
Tips and checklists to help with evaluating AI vendors

From AI skeptics to believers to enthusiasts, this guide provides tactical advice on what safe and successful AI adoption looks like — and how your security team can accomplish it.

Get the Securing AI in the Enterprise guide today to start making real progress toward achieving your AI adoption goals.

Source: OpenAI

OpenAI is reportedly in talks to raise an additional $40 billion in venture funding at a $340 billion valuation. The talks were first reported by the Wall Street Journal. OpenAI didn’t return a request for comment regarding the report.

This would, by a long shot, become the largest single venture funding round ever.

The details: The round would be led by Softbank, which would contribute between $15 billion and $25 billion. Softbank recently partnered up with OpenAI on Project Stargate.

It’s not clear how the funding will be used, but it could well be put toward some of OpenAI’s financial commitments related to Project Stargate; OpenAI said during the Stargate announcement that it would commit $19 billion to the $500 billion effort.
Just a few months ago, OpenAI raised $6.6 billion in venture funding at a $157 billion valuation, making it one of the largest (by valuation) private companies in the world. The startup at the same time secured a $4 billion revolving line of credit.

This is in addition, of course, to the $13 billion + that Microsoft has poured into the firm over the past few years. Roughly one year ago, OpenAI was valued at $80 billion, a figure that was eye-popping then.

OpenAI reportedly lost about $5 billion in 2024 on nearly $4 billion in revenue, and expects these losses to continue spiking until the end of the decade. The talks rather conspicuously come shortly following the DeepSeek panic, in which many investors, policymakers and technologists worried that the business of AI — hundreds of billions worth of compute, with little revenue to show for it — is no longer as strong as it once appeared.

The Vatican (kind of surprisingly?) published a lengthy exploration of the distinctions between human and artificial intelligence: “AI’s advanced features give it sophisticated abilities to perform tasks, but not the ability to think.”
Apple reported revenue of $124.3 billion for the quarter, though it noted an 11% sales decline in China. The stock moved slightly lower in extended trading.

Exposed DeepSeek database revealed chat prompts and internal data (Wired).
US economy finished 2024 on solid footing, as tariff threats weigh on 2025 forecasts (Semafor).
An AI chatbot in Paraguay is trying to humanize prisoners (Rest of World).
Anthropic CEO says limiting China’s access to AI chips is ‘existentially important’ (404 Media).
Is Tesla still a car company? (The Verge).

Secure elite creative professionals from Latin America to scale your team.
Save 70% on salaries with Athyna’s global talent platform.
Enjoy a $1,000 discount exclusively for readers of The Deep View.

Find Out More! *

Source: Los Alamos

OpenAI said Thursday that it has secured a partnership with the U.S. National Laboratories to “supercharge” the labs’ scientific advancements using OpenAI’s generative technology.

The details: The Energy Department operates 17 National Labs, each of which focuses on a distinct discipline. This, perhaps most notably, includes the Los Alamos lab, which once served as the site of Oppenheimer’s Manhattan Project.

OpenAI said it will work with Microsoft to deploy o1 on Venado, Los Alamos’ Nvidia-sourced supercomputer. The Venado machine will operate on a secure network, where it will be a shared resource for several additional labs.
Los Alamos said in a statement that nearly all of the labs’ missions are already leveraging AI in some capacity. The lab detailed some of the ways in which OpenAI’s AI might help, including research into disease treatment and prevention, updating the country’s energy infrastructure, protecting the power grid and enhancing national security through early threat detection. But the specifics of how o1 will be applied to these arenas remain unknown.

The model will also apparently aid the Labs’ efforts in enhancing nuclear security, something that will be supported “with careful and selective review of use cases and consultations on AI safety from OpenAI researchers with security clearances.”

The Department of Energy did not return a request for comment on the specifics and details of those use cases.

Los Alamos has previously collaborated with OpenAI in efforts to improve model safety.

The financial terms of the partnership remain unknown. This latest move further increases OpenAI’s governmental ties, shortly following its announcement of ChatGPT Gov.

The broader landscape: Such partnerships are not unusual; Microsoft, IBM and Nvidia (plus plenty of universities) are partnered with, or have partnered with, National Labs.

Source: Unsplash

The debate on AI and copyright involves a couple of moving parts. One is whether AI-generated content is eligible for copyright protection, and the other — far more significant one — involves a simple question of whether it is ‘fair use’ for developers to train their models on copyrighted content (without authors’ permission).

In its first movement on the issue in months, the U.S. Copyright Office this week released the second of three reports on artificial intelligence. The first, issued last year, called attention to the problem of digital replicas and deepfakes. The third and final part of this report will tackle the fairness of training AI models on copyrighted material.

This second part answers the question of whether AI-generated material is eligible for copyright protections.

Here’s the gist: The 41-page report stated quite clearly that purely AI-generated material — “or material where there is insufficient human control over the expressive elements” — cannot be copyrighted.

But, if a human uses generative AI to assist in their creative endeavors, the end result is eligible for copyright protection. In determining the right level of human authorship, “prompts do not alone provide sufficient control.” Part of the reasoning for this is that the same prompt can generate a number of distinct outputs, an example of an “AI system providing varying interpretations of the user’s directions.”
The report said that no new legislation is needed here; the significance of human contributions to AI-generated works can be settled on a case-by-case basis.

The distinction here, according to the office, depends not on the system, but on how the system is used. Leveraging AI to de-age an actor, for instance, or to remove unwanted objects from a scene, doesn’t impact the copyrightability of a work; “using AI as a stand-in for human creativity” does.

The Office rejected an argument for applying additional legal protection to AI-generated material, saying that copyright requires human authorship.

“We share the concerns expressed about the impact of AI-generated material on human authors and the value that their creative expression provides to society,” the report reads. “If authors cannot make a living from their craft, they are likely to produce fewer works. And in our view, society would be poorer if the sparks of human creativity become fewer or dimmer.”

Underpinning all of this is a reality in which, seemingly due to generative AI, jobs for artists — specifically freelance digital artists, graphic designers, writers, video game designers, computer coders — are drying up. Researchers last year identified a 30% decrease in the availability of freelance writing jobs, a 20% decrease in the availability of freelance coding jobs and a 17% decrease in demand for graphic designers.

This seems to clear up Hollywood’s cautious, partial adoption of the tech.

Largely, this was an affirmation of the Copyright Office’s initial stance on AI-generated content; the thing that’s still missing is guidance on whether training constitutes ‘fair use.’ The fact that we still don’t have it is a good indicator that answering that question is neither clear nor straightforward.

It’s not clear when it’s coming, but when it comes, that third part will be a bombshell for the industry, regardless of what side it lands on. And the impacts of that will be far more significant to the careers of artists than the copyrightability of AI-generated content.