Defining a single source of truth with triple entry accounting: UNISOT’s Torje Sunde on CoinGeek Backstage

In his paper on Triple Entry Accounting (TEA), Torje Sunde delved into how connecting modern accounting ledgers to a public blockchain would make auditing easily verifiable. In an interview with CoinGeek Backstage in Malta, he discussed how the paper impacts auditing and curbs fraud.

Sunde was one of the presenters at the inaugural Triple Entry Accounting Conference in November 2023. Held in Malta, the event was an initiative of industry titan Ian Grigg and brought together accounting professionals, tech gurus, blockchain experts, and more to discuss the intersection of accounting and public blockchain.

Sunde’s paper, authored alongside Dr. Craig Wright, is titled “Implementing Triple Entry Accounting as an Audit Tool—An Extension to Modern Accounting Systems.”

“One of the key objectives of the paper is to attract people from different disciplines: accountants, developers, experts in mathematics and cryptography. We need them all to come together to build this triple entry accounting system,” he told CoinGeek Backstage.

Sunde, the CFO at Norwegian blockchain supply chain management firm UNISOT, believes TEA is a game-changer for auditors. In the paper, he outlines a method of verifying accounting ledgers and proving that only a single ledger exists.

“With the current double-entry accounting, people can have multiple sets of books. They can show one book to someone, change the numbers, and show fictitious numbers to someone else. With triple entry accounting, we can define a single source of truth that can be verified,” he added.

Triple Entry Accounting in AI

Besides financial auditing, TEA has massive applications in artificial intelligence (AI),
Konstantinos Sgantzos points out.

Sgantzos heads the environmental sub-sector at the Public Power Corporation S.A. (NASDAQ: PUPOF), the largest grid operator in Greece. He is also a researcher at the
University of Sharjah in the United Arab Emirates with an interest in AI and the large language models (LLMs) that underpin them.

At the conference, Sgantzos’s presentation centered on applying TEA to protect intellectual property (IP) in AI training. As the technology takes root, one of the most contentious topics has been around the use of IP.

Almost every major AI developer faces dozens of lawsuits from artists, news organizations, authors, and other creatives whose work has been used to train LLMs without their knowledge or consent. The New York Times sued OpenAI and Microsoft (NASDAQ: MSFT) over accusations of copyright infringement in a lawsuit it said was worth billions of dollars last December. OpenAI was sued for $3 billion for alleged illegal internet data scraping earlier last year.

Sgantzos says that with TEA, content creators can finally have control over who uses their content and, where applicable, get adequately compensated.

“The proposition is that each contributor can use this system of triple entry accounting [and] a cryptographic proof so that they are aware that their content is being used,” he told CoinGeek Backstage.

For instance, an artist can embed a cryptographic signature into his art, allowing him to expressly consent every time an LLM system seeks to use that data to train.

Currently, models like DALL-E and Stable Diffusion can create art in seconds, leveraging millions of images of real art that took hours to make. Data shows that since 2022, AI users have created over 15 billion images using existing AI tools. Before AI, it took humanity 150 years to produce as many images.

In order for artificial intelligence (AI) to work right within the law and thrive in the face of growing challenges, it needs to integrate an enterprise blockchain system that ensures data input quality and ownership—allowing it to keep data safe while also guaranteeing the immutability of data. Check out CoinGeek’s coverage on this emerging tech to learn more why Enterprise blockchain will be the backbone of AI.

Watch: The Bitcoin Masterclasses—Making accounting sexy again