Press Room

Dear NCA Members: 

In May, Informa, the parent company of NCA’s publishing partner Taylor & Francis (T&F), announced a $10 million data-access agreement with Microsoft for “nonexclusive access to Advanced Learning Content” across its books and journals. 

We want to start by apologizing for how long it has taken for the Executive Committee of the Legislative Assembly to communicate directly to you about this issue. We have been working on making sure we understood the meaning of Informa’s sale of access to NCA publications for AI as well as the impact on NCA’s membership. Quite frankly, it has taken this long for us to get the fullest details we can from Taylor & Francis. 

The Executive Committee was not consulted about or given advance notice of this deal. The Publications Council Chair and EC member, John M. Sloop, immediately contacted T&F to understand the implications of this deal for our journals and to ask why NCA had not been consulted. In its initial response, T&F indicated that NCA’s publications (journals and books) were not included in the Microsoft deal. Subsequently, NCA’s principal contact at T&F informed NCA that this information was incorrect due to internal miscommunication in their organization. As of now, T&F has stated that NCA journal content is being made available to Microsoft for purposes of Artificial Intelligence (AI) training purposes only (and that NCA journal content otherwise is not being republished). As part of NCA’s multiyear contract with T&F, each author agrees to grant T&F the legal right to make NCA publications available. These are the standard procedures followed by all major publishers.   

Taylor & Francis has provided the EC with both a series of talking points and further background information that outline the philosophy that guides opportunities like this and the entire world of Large Language Models. You can review those T&F talking points and T&F background information below. NCA annually receives a large amount of royalties from T&F for NCA journal content and has been promised some level of financial compensation from Informa’s Microsoft contract—although T&F so far has not told NCA how much or when. 

Many EC members are published scholars who share your concerns and questions.  One major concern is that the EC and NCA staff were unaware of this sale until after it was announced publicly. We acknowledge that T&F and its parent company were not under any obligation to advise of this sale. However, we request that T&F, as our publishing partners, in the future inform us and seek our counsel and advice regarding agreements affecting NCA publications. Those who publish in NCA journals should be aware of how their work might be used and are consulted about potential uses. In addition, NCA will engage in discussions with similar academic organizations that also are T&F clients to identify possible shared positions and goals. 

Finally, we are continuing our conversations with T&F on this—and other—issues relevant to academic integrity. As always, we are open to your input. We strongly encourage you to contact NCAEC@natcom.org with any questions or concerns. 

Sincerely,  

NCA Executive Committee of the Legislative Assembly 

Marnel Niles Goins, President 

Jeanetta D. Sims, First Vice President 

Tina M. Harris, Second Vice President 

Walid Afifi, Immediate Past President 

Jimmie Manning, Finance Committee Chair 

Justin Danowski, Executive Director 

Candice Thomas-Maddox, Finance Committee Director 

Kenneth Lachlan, Finance Committee Director 

James L. Cherney, IDEA Council Chair 

Laurie Lewis, Mentorship and Leadership Council Chair 

John M. Sloop, Publications Council Chair 

Shaunak Sastry, Research Council Chair 

Katherine S. Thweatt, Teaching and Learning Council Chair 

“Talking Points” Statement from Taylor & Francis to NCA 

“With regards to AI in general: 

• We believe that academic knowledge has a fundamental role to play in supporting the improvement, relevance, and performance of AI models that are used both by our academic stakeholders and the world at large. 
• Our role as a publisher is to help authors’ ideas make their fullest contribution to society. The work we are doing with AI will improve research by making it easier to analyse data, generate hypotheses, automate tasks, work across disciplines, and research ideas. 
• We believe it is far better that Large Language Models (LLMs) are trained on trusted, substantiated academic knowledge. 
• Licensing is a fundamental part of what we do as an academic publisher, it’s part of our commitment to authors. 
• We are entering into agreements with partners who are aligned with Taylor & Francis on the importance of accurate citation and author rights 
• We have contractual rules that protect authors rights, safeguard their content from unauthorized access or use, limit the reproduction of verbatim text, and ensure royalties are paid appropriately. 
• Royalties will absolutely be paid to authors and other rightsholders in accordance with the licensing terms and royalty statement periods in their contracts. 

“In case this is useful for the Exec Committee to understand, as we carefully work through these opportunities, we are guided by four principles: 

• Responsible partners who are aligned with Taylor & Francis on the importance of accurate citation, of author rights, and of improving the impact of scholarly publications. 
• Protection of content and IP through contractual rules that protect authors and other rightsholders and safeguard their content from unauthorized access or use. 
• Safeguards that limit the reproduction of verbatim text extracts in outputs. 
• Payment of royalties to authors and other rightsholders in accordance with their contracts. 

“The rationale for these deals is that licensing is an everyday part of what we do for our authors. And we are excited by the opportunity offered by bringing together the world’s most advanced thinking with AI. We believe that academic knowledge has a fundamental role in supporting the improvement, relevance, and performance of AI models that are used both by our academic stakeholders and the world at large. 

“Specifically, there is opportunity to: 

• Improve the accuracy and trustworthiness of AI by training LLMs and expert agents on trusted, substantiated, academic knowledge. 
• Improve research by making it easier to analyse data, generate hypotheses, automate tasks, work across disciplines, and research ideas. 
• Enhance services for authors and readers e.g. automated literature reviews, Plain Language Summaries, translation, and educational tools. 

“Background” information statement from Taylor & Francis: 

“…Taylor & Francis have been investing in machine learning, automation and other AI technologies for many years. We use AI to improve research submission, review, production, publication and distribution for the benefit of authors, editors and institutions. Some of those ways include automated screening that accelerates the acceptance of research; matching articles to journals to help find the right home for new research; providing first line automated plagiarism checks to maintain research integrity; improving efficiency in the production process to reduce time to publication; and tagging content and applying meta data automatically so it becomes easier to discover, use and share. 

“Our starting point is our responsibility to assess the role AI could and should play in what we do: which is serving the research and academic community and making sure expert knowledge can make its fullest possible contribution to society. Our assessment is that there are significant and meaningful opportunities for our community from the further development of AI. And so, we have chosen to engage with AI by forming partnerships in a carefully considered way. 

“These partnerships enable us to contribute to the discussion, development and design of AI models while learning more about the technology and its application as it evolves. We have found partners who are aligned with us on the importance of knowledge and original research, the importance of verification, authenticity and trust, and the importance of respecting intellectual property rights. 

“Our partnerships have several specific aspects. We are exploring the development of specialised expert agents and Small Language Models (SLMs). These new AI products for the research community have the potential to further help the discovery of knowledge, impact of research and effectiveness of learning. We are also working on research-specific tools such as automated citation technology. While it is early days, these are potentially very exciting enhancements and products that could materially expand the reach and impact of research. 

“In addition, we are providing data and content under licence for the purposes of training AI, such as LLMs, so that those models become more accurate and relevant for the benefit of everyone who uses them…. 

“There are strict boundaries attached to our partnership agreements. Data and content are being used for training purposes only, and under no circumstances can they be reproduced in an equivalent format. The agreements include full protection of intellectual property rights and authors’ rights. 

“These partnerships will generate a profit, after the payment of tax. A considerable proportion of those proceeds will rightly return to authors, societies, and other rightsholders, in keeping with their existing contracts and royalty and licencing agreements. NCA will therefore receive royalties in accordance with our existing agreement. As publishing partner, it is our responsibility under the terms of our agreement to optimize opportunities for the NCA content such as these and to support and enhance financial returns for the Association.   

“A further proportion – one third – will be reinvested back into our product and technology development so that we can keep improving what we deliver and develop future research services. “At a minimum, those areas include the further expansion of our open research offer to deliver more choice to researchers and institutions; enhancing the technology platforms that deliver research to keep improving discoverability and adding new value-add features; and keeping up the pace of our programmes around research integrity. 

“These AI partnership reinvestments are in addition to the ongoing investment that T&F makes in the knowledge community. This includes investing in shared infrastructure to guard against the huge increases in research integrity issues and piracy, such as supporting the STM Integrity Hub and the need for AI detection tools. 

“We recognise that the rapid rise of AI has generated a range of strong opinions. Our perspective is that academic knowledge has a fundamental role to play in the improvement, relevance, and performance of AI models that are used both by our academic stakeholders and the world at large, and it is only by engaging proactively with AI developers that we will achieve the best outcome for our authors and society partners. This is precisely why we have put in place contractual rules that protect authors’ rights, safeguard their content from unauthorized access or use, and limit the reproduction of verbatim text.”