Between Copyright and Computer Science: The Law and Ethics of Generative AI

Deven R. Desai; Mark Riedl

doi:10.2139/ssrn.4735776

journal article Jan 01, 2024

Between Copyright and Computer Science: The Law and Ethics of Generative AI

Deven R. Desai Mark Riedl

SSRN Electronic Journal · Elsevier BV

View at Publisher Save 10.2139/ssrn.4735776

Topics

No keywords indexed for this article. Browse by subject →

References

77

[1]

See Radford "Language Models Are Unsupervised Multitask Learners" OPENAI (2019)

[2]

The smallest model [summarized in Table 2 with 117M parameters] is equivalent to the original GPT

[3]

"With 1.5 trillion parameters compared to GPT-3.5's 175 billion, GPT-4 clearly outperforms GPT-3.5 in terms of model size and parameters" GPT-3.5 and GPT-4 Comparison: Exploring the Developments in AI-Language Models, MEDIUM (2019)

[4]

Mary Harshita Anthropic Forecasts More than $850 Mln in Annualized Revenue Rate by 2024-End-Report (2023)

[5]

Appenzeller supra note 80 (noting extreme expense to train models but that such costs are manageable for a "well-funded start-up

[6]

are going to respect and support the individuality of members, then a requisite may be that we know a great deal about them

[7]

R Accord Deven "Exploration and Exploitation: An Essay on (Machine) Learning, Algorithms, and Information Provision, 47 LOY. U. CHI" L.J (2015)

[8]

See Laura Chronology of Twentieth-Century Science, in A CENTURY OF NATURE: TWENTY-ONE DISCOVERIES THAT CHANGED SCIENCE AND THE WORLD xv, xv-xviii (2003)

[9]

Id 116 Campbell v (1994)

[10]

Id (1998)

[11]

lack of permission is 'beside the point (1991)

[12]

See Kyle How the New York Times Is Using Generative AI as a Reporting Tool, ARS TECHNICA (2024)

[13]

See Alexandra Inside the Movement Behind Trump's Election Lies (2024)

[14]

E G See When the New York Times Lost Its Way, The ECONOMIST (2023)

[15]

N Y See

[16]

N Y The Upshot TIMES

[17]

Natalie Gil New York Times Launches Data Journalism Site The Upshot, GUARDIAN (2014)

[18]

N Y Times The Upshot Staff, 10 Data Points and Documents that Made Us (2023)

[19]

The 6 Kinds of Republican Voters (2023)

[20]

Id "Clustering is a central task for which many algorithms have been proposed" ACM SIGKDD EXPLORATIONS NEWSLETTER (2002)

[21]

"The IMDb Services or any portion of such services may not be reproduced, duplicated, copied, sold, resold, visited, or otherwise exploited for any commercial purpose without express written consent of IMDb" 175 IMDb Conditions of Use

[22]

See Parlapiano, supra note 174

[23]

Aws Marketplace

[24]

Cf

[25]

Josef Adalian "And then there's what's often the biggest obstacle to getting a show on to a streaming platform: music rights" Why Are These Classic Shows Nowhere to Be Found on Streaming?, VULTURE (2016)

[26]

Id

[27]

E G See Regarding Unreleased Doc Accusing Musician of Abuse, THE HOLLYWOOD REP (2024)

[28]

See Michael "The Tragedy of the Anticommons: Property in the Transition from Marx to Markets, 111 HARV" L. REV (1998)

[29]

A Michael Can Patents Deter Innovation? The Anticommons in Biomedical Research, 280 SCI (1998)

[30]

See Heller

[31]

H Jerome "When Copyright Law and Science Collide: Empowering Digitally Integrated Research Methods on a Global Scale, 96 MINN" L. REV (2012)

[32]

E G See "OpenAI Strikes Content Deal with Cond� Nast, Raising Questions About Future of Publishing" VENTUREBEAT (2024)

[33]

Benj Bellan Deeply Troubled" by OpenAI's Content Deals with Vox, The Atlantic, ARS TECHNICA (2024)

[34]

E G See A Devil's Bargain with OpenAI, THE ATLANTIC (2024)

[35]

E G See

[36]

Ftc (2023)

[37]

Beres, supra note 187 (noting the issues around AI as a threat to writers and the way previous deals between technology companies and journalism ended up not benefitting the news outlet

[38]

Nu�ez supra note 186; Edwards & Belanger, supra note 186

[39]

Kyle Wiggers ] (noting the growing need for more data to train LLM models and the multi-billion dollar market emerging to meet that demand) (2024)

[40]

D Nev Samuelson, supra note 5 (1106)

[41]

Grp., 989 F.3d 1205 (2021)

[42]

"The mere threat of lawsuits and high damages has a history of preventing action, especially when intellectual property is at stake" Deven R. Desai, Speech, Citizenry, and the Market: A Corporate Public Figure Doctrine, 98 MINN. L. REV (2013)

[43]

Cf (1190)

[44]

E G See Real-World Use Cases for Large Language Models (LLMs) (2023)

[45]

See Mccarthy

[46]

Justine Calma Twitter Just Closed the Book on Academic Research, THE VERGE (2023)

[47]

Id [E]ven its 'outrageously expensive' enterprise tier, the coalition argued, wasn't enough to conduct some ambitious studies or maintain important tools

[48]

E G See

[49]

Id at 210 (Google's program does not allow a searcher to increase the number of snippets revealed by repeated entry of the same search term or by entering searches from different computers. A searcher can view more than three snippets of a book by entering additional searches for different terms. However, Google makes permanently unavailable for snippet view one snippet on each page and one complete page out of every ten-a process Google calls "blacklisting

[50]

See supra Part III.A (discussing how academic benchmark goals differ from commercial ones)

Showing 50 of 77 references

Metrics

5

Citations

77

References

Details

Published: Jan 01, 2024

Authors

Cite This Article

Deven R. Desai, Mark Riedl (2024). Between Copyright and Computer Science: The Law and Ethics of Generative AI. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.4735776

Between Copyright and Computer Science: The Law and Ethics of Generative AI

You May Also Like