Elsevier's Chi urges focus and trust as AI pressures universities
Elsevier chairman Youngsuk “YS” Chi told university leaders that AI governance lags expectations, then used the 20-year lithium-ion battery story to argue for focus, quality and trusted knowledge.

Why it matters
- Elsevier surveyed 450 university leaders in 20 countries; most rated all 25 institutional goals as priorities.
- More than two-thirds expect AI to drive new knowledge within 2 to 3 years, but only about one-third report good AI governance.
- The lithium-ion battery took about 20 years and a partnership spanning three countries to become a product.
- John Goodenough, M. Stanley Whittingham and Akira Yoshino shared the 2019 Nobel Prize in Chemistry.
- Korean companies supply more than one-quarter of electric vehicle batteries used outside China, Chi said.
By Elena Vasquez
At Korea University in Seoul on Wednesday, Youngsuk “YS” Chi said only about one-third of academic leaders have good AI governance, while more than two-thirds expect AI to drive new knowledge within two to three years.
Chi’s contrast sets the central policy challenge for universities: how to use AI quickly without building new systems on weak governance or untrusted evidence.
Chi, chairman of Dutch academic publisher Elsevier, delivered the keynote on the second day of the Innovation & Impact Summit 2026 at Korea University.
Times Higher Education (THE), the London-based publication known for its annual World University Rankings, is holding the summit from Tuesday through Thursday.
His address began with pressures that Chi said university leaders could no longer separate.
Chi said: “Many of you are facing tighter budgets. Geopolitical tension is making it harder to work across borders. AI is changing how research is done faster than anyone can write the rules about it.”
He added: “None of these pressures is new on its own. What is new is that they are arriving at the same time.”
For university leaders, that convergence changes what responsible execution requires. Each pressure consumes attention, while none offers a simple stand-alone solution.
His response was not to promise certainty. He asked institutions to choose, partner, measure and explain value while the ground keeps moving.
The policy and research stakes are immediate. Leaders must set priorities, protect cross-border work and govern AI without waiting for settled rules.
What does the battery story teach about focus?
Chi built the main argument around the lithium-ion battery. Its developers, John Goodenough, M. Stanley Whittingham and Akira Yoshino, shared the 2019 Nobel Prize in Chemistry for their work.
Goodenough, then 97, made a key breakthrough at the University of Oxford, according to Chi’s account.
Chi said the university declined to patent the discovery. It captured almost none of its commercial value, while “the whole world gained.”
The battery history separates invention, commercialization and public acceptance. Oxford supplied the key breakthrough. The three-country partnership helped turn it into a product. Yoshino’s work on safety helped earn trust.
The 2019 Nobel recognition came after a long path from laboratory work to a product people carried every day. Chi used that span to challenge shorter institutional timelines.
He used that outcome to separate effort from optimism.
“Hard isn’t bad. Hard is just hard,” Chi said. “Optimism without thoughtful action is just wishful thinking.”
His point was practical. A difficult research problem still needs patient work, clear priorities and institutions able to sustain a long path.
How are universities setting their priorities?
Chi then set out five questions for university leaders.
- Where should the institution focus?
- Who should it work with?
- Can it carry ideas forward?
- How will it know whether its work is working?
- How will it show the work’s value?
Together, the questions shift the discussion from ambition to execution. They ask leaders to identify purpose, build durable relationships and prove results.
The five questions also form a sequence. Institutions first choose a direction, then secure partners and capacity, then measure performance and communicate value.
That order matters because universities cannot treat every objective as equally urgent. They must still decide which work deserves sustained attention and which commitments require a clearer sequence.
His evidence came from an Elsevier survey of 450 university leaders in 20 countries. Because the survey spans 20 countries, Chi’s priority problem appears broad rather than confined to one national system.
Most respondents rated every one of 25 institutional goals as a priority.
“When everything is a priority, nothing is,” Chi said.
That finding describes a broad appetite for institutional improvement, not a usable order of operations. Leaders can support many ambitions without giving each one equal time, money or attention.
Chi returned to the battery to explain why focus requires more than choosing a short list. The battery emerged from an oil company, a university and a chemical company in three countries.
He said the depth of that partnership mattered more than the number of partners. The three institutions had to connect different capabilities across national borders and sustain that work over time.
The discovery also took about 20 years to become a product.
That 20-year timeline gives Chi’s first question a concrete measure. Focus must survive delays, technical setbacks and the need to keep several partners aligned.
Chi said Yoshino’s key contribution was battery safety. That work enabled billions of people to trust the batteries in their pockets.
“Quality over quantity is not a slogan. It is how public trust is built,” he said.
The battery history gives those priorities a testable meaning. Safety, useful products and public confidence matter more than a long list of partnerships or promises.
What does AI change for research governance?
Chi returned to AI with another sharp contrast. More than two-thirds of academic leaders believe it will be the main driver of new knowledge within two to three years.
Only about a third, however, report good AI governance at their institutions.
The first measure captures expectation. The second captures institutional readiness.
Chi tied the gap to the quality of research infrastructure. His timing also matters. AI is changing research faster than universities can write the rules, while governance requires deliberate rules, oversight and institutional ownership.
“AI is only as good as the knowledge it is built on,” he said. “It’s not going to be about the biggest model size. It’s going to be about the trusted model.”
For research leaders, the distinction moves attention from model scale alone toward the records, methods and governance supporting each system. Institutions must first know whether their knowledge foundations can support reliable outputs.
That matters because trusted knowledge, rather than the largest model, forms the basis for decisions about discovery, policy and public use. It also gives research institutions a reason to treat quality control as core infrastructure rather than an afterthought.
Chi’s warning was not that universities should avoid AI. It was that they should not mistake rapid adoption for institutional control. The governance gap means many leaders face new tools before they have finished defining the standards around them.
Can trusted knowledge still scale?
Chi also used the battery to connect trusted knowledge with industrial scale.
Korean companies helped take the technology to global scale, Chi said. They still supply more than a quarter of the electric vehicle batteries used outside China.
He presented that record as evidence for focused partnerships and sustained execution, not as a standalone industrial statistic. Global scale followed years of work in which safety, useful products and public confidence mattered alongside scientific breakthroughs.
Elsevier’s position gives his argument a publishing context. The Dutch academic publisher was founded in 1880 and publishes scientific journals including The Lancet and Cell.
A publisher depends on reliable knowledge because errors and weak controls travel quickly through shared research. That makes quality and trust central to the business, not ethical decoration.
Chi’s speech closed by joining those two concerns. Universities need trusted knowledge, and AI needs governance strong enough to preserve that trust.
The Innovation & Impact Summit 2026 continues through Thursday. University leaders now have to answer Chi’s five questions with fewer priorities, deeper partnerships, credible AI governance and evidence that their work creates value.
Source: feed.koreatimes.co.kr
More from Elena Vasquez
Show full bio
Market editor covering technology and product news in artificial intelligence and machine learning at AI In Context.
708 articles
Related articles
- 95 Percent of Organizations See No Measurable AI Gains, University Leaders Warn
- Over 300 Delegates From 40 Countries Gather in Seoul for THE Innovation & Impact Summit 2026
- Google DeepMind Partners with Korea to Bring Frontier AI to Science
- Stanford's AI100 Report: Scaling Era Hits Data and Cost Walls
- Universities Must Become 'Neutral Brokers' as AI Forces Mission Rethink