New report highlights the potential for artificial intelligence to accelerate the real-world impact of research
Taylor & Francis Group
A new report by HEPI and Taylor & Francis explores the potential of AI to advance translational research and accelerate the journey from scientific discovery to real-world application.
Using Artificial Intelligence (AI) to Advance Translational Research (HEPI Policy Note 67), authored by Rose Stephenson, Director of Policy and Strategy at HEPI, and Lan Murdock, Senior Corporate Communications Manager at Taylor & Francis, draws on discussions at a roundtable of higher education leaders, researchers, AI innovators and funders, as well as a range of research case studies, to evaluate the future role of AI in translational research.
Key findings
The report finds that AI has the potential to strengthen the UK’s translational research system, but that realising these benefits will require careful implementation, appropriate governance and sustained investment.
Key findings include:
- AI could accelerate translational research by enabling faster analysis of large and complex datasets, supporting knowledge synthesis and improving links between disciplines. However, the availability and quality of such datasets remain uneven, limiting the ability of AI tools to support research translation in some fields.
- Access to AI skills and expertise is increasingly important and building this access into interdisciplinary frameworks will be a key component of driving translational research.
- AI can improve the accessibility and visibility of research, including through plain-language summaries, semantic search (search functions that utilise concepts and ideas and not simply keywords, giving a more accurate result) and new formats aimed at audiences beyond academia.
- There are clear risks associated with AI use, including challenges around reproducibility, bias, deskilling, academic integrity, intellectual property and accountability.
Recommendations
To ensure AI supports high-quality and responsible translational research, the report makes recommendations for research funders, institutions and publishers, including:
- Setting clear expectations for the responsible use of AI, including alignment with guidance such as the UK Research Integrity Office’s Embracing AI with Integrity.
- Investing in trustworthy and ethical AI, including work to improve transparency, reduce bias and support reproducibility.
- Strengthening support for interdisciplinary research, including better recognition of team-based work and clearer routes to access AI expertise.
- Supporting shared and open AI research infrastructure to reduce duplication and make researcher-developed tools more widely available.
- Encouraging data sharing and reuse, alongside investment in infrastructure that supports secure and responsible access to data.
Rose Stephenson, Director of Policy and Strategy at HEPI and co-author of the report, said: “The UK has extraordinary research strengths, but too many ideas struggle to make the journey from discovery to real-world use. AI has the potential to support this process by speeding up analysis, connecting disciplines and improving access to research. However, these benefits will only be realised if AI is used transparently, ethically and in ways that strengthen, rather than replace, human expertise.”
Rebecca Lawrence, VP Knowledge Translation at Taylor & Francis, said: “We are grateful to all the roundtable participants and those who shared case study insights. The valuable discussions and the ensuing process of putting the policy note together has highlighted the benefits of working collectively to harness the power and opportunity that responsible AI use can provide for translational research.”
By investing in interdisciplinary expertise, ethical governance and infrastructure, stakeholders can help transform translational research, enabling more of the latest research to deliver meaningful societal benefits.
Read the full report: Using Artificial Intelligence to Advance Translational Research
Using the physics of radio waves to empower smarter edge devices
Duke engineers publish new method to use analog radio waves to boost energy-efficient edge AI
Duke University
As drones survey forests, robots navigate warehouses and sensors monitor city streets, more of the world’s decision-making is occurring autonomously on the edge—on the small devices that gather information at the ends of much larger networks.
But making that shift to edge computing is harder than it seems. Although artificial intelligence (AI) models continue to grow larger and smarter, the hardware inside these devices remains tiny.
Engineers typically have two options, neither ideal. Storing an entire AI model on the device requires significant memory, data movement and computing power that drains batteries. Offloading the model to the cloud avoids those hardware constraints, but the back-and-forth introduces lag, burns energy and presents security risks.
Researchers at Duke University are exploring a third option, called WIreless Smart Edge networks (WISE), that bypasses the limitations of both approaches. They’ve shown that large AI model weights can be smartly embedded in the form of radio waves delivered over the air between devices and nearby base stations, opening a path to energy-efficient edge AI without the usual cost in energy, speed or size.
This work, published online in Science Advances on January 9, is led by Tingjun Chen, the Nortel Networks Assistant Professor of Electrical and Computer Engineering, alongside Dirk Englund’s team at the MIT Research Laboratory of Electronics (RLE). This work was supported by the NSF Athena AI Institute, with subsequent continuation and expansion also supported by the Army Research Office.
At the heart of the approach is a concept called in-physics analog computing.
Traditional digital computing occurs through binary code. Devices convert data into ones and zeros, move those bits into a digital processor and compute long sequences of math operations. Even a simple task like unlocking a phone with biometrics triggers a rapid sequence of calculations. It’s reliable but not efficient for small, battery-powered devices.
In-physics computing works differently. Instead of shuttling ones and zeros from an edge device to a distant processor, the natural behavior of radio waves completes part of the math along the way.
In WISE, a base station stores the full AI model and broadcasts a radio frequency (RF) signal that encodes the model’s weight values—numbers required to complete those calculations. When the signal reaches a nearby device, radio hardware in the device mixes the broadcast signal with its own input data that can naturally perform computing directly in the RF or analog domain. One example is a passive frequency mixer that “approximates” the multiplication of two time-domain RF signals. That analog in-physics mixing process—directly taking place at RF—performs a key step in most deep learning models without the need of a digital processor.
“We’re taking advantage of computations that common, miniaturized electronics already gives us,” Chen said. “Instead of running every step of the model on a chip designed for digital computing, the radio waves themselves help carry information in a way optimized for the computation.”
Because the device doesn’t store the entire model or run it digitally, it overcomes the big memory and energy costs that limit edge AI today.
Zhihui Gao, a PhD student in Chen’s lab and lead author on the paper, said the idea could benefit many kinds of devices. Drones, cameras and traffic sensors all generate data continuously, yet they struggle to run the advanced models that would help interpret those data.
“Technology is moving toward smaller devices that can do more than ever before,” Gao said. “In order to achieve that, we need new improvements in edge computing. With WISE, we have shown how devices can run on powerful AI without relying on heavy chips or distant servers.”
Gao noted another advantage of WISE is its ability to use existing infrastructure. Base stations already set up for 5G, emerging 6G or WiFi routers could be augmented to broadcast these AI models with relatively small adjustments. Plus, everyday wireless devices already contain the hardware, such as frequency mixers, needed to perform the in-physics computation.
“We’re not adding exotic components or creating entirely new hardware,” Gao said. “We’re reusing features that are widely deployed and don’t consume extra energy.”
In experiments, WISE achieved nearly 96 percent image classification accuracy while consuming more than an order of magnitude less energy than leading digital processors.
Although promising, WISE is still in its beginning stages. The current prototype works over short distances, but longer-range testing would require stronger transmission or integration with next-generation wireless gear. And while the approach is flexible, broadcasting multiple AI models simultaneously would require efficient multiplexing of the time-frequency-space resources or additional spectrum bandwidth.
Even so, the researchers see broad potential in applications. One base station could support a swarm of drones in a search and rescue mission or help traffic cameras coordinate intersection signals.
“This is the next step in wireless technologies becoming as powerful as wired ones,” Chen said. “Beyond delivering data and information, these findings open a new direction, in which future networks may distribute intelligence by blending communication and computation to enable energy-efficient edge AI at massive scale.”
Journal
Science Advances
Method of Research
Experimental study
Subject of Research
Not applicable
Article Title
Disaggregated machine learning via in-physics computing at radio frequency
Article Publication Date
9-Jan-2026
AI-everything? Consumer groups name the
worst tech products at CES 2026
An AI-powered fridge, electronic lollipops, and Amazon’s Ring cameras were named by consumer groups as some of the worst products at this year’s tech show.
Barely a week into January, artificial intelligence (AI) is already the defining feature of this year’s consumer technology landscape.
At CES 2026, the annual gadget fair that ended on Friday in Las Vegas, AI was everywhere. But critics and consumer protection groups are questioning whether too many products are being labelled “intelligent” without good reason.
“Worst in Show” is an annual contest that highlights what organisers see as the most wasteful technology on display at the show. It is judged by a panel of consumer and privacy advocates, including US nonprofit Consumer Reports, refurbished tech marketplace Back Market, and right-to-repair advocates iFixit.
Organisers published their list of losers on Thursday, saying they had chosen “the most overengineered, unrepairable and unsustainable technology disasters at CES” this year.
Topping the list was Samsung’s Bespoke AI Family Hub, an internet-connected fridge that can open and close its door using voice commands.
Judges said the voice controls, constant internet connection, and built-in advertising could make what’s normally a simple household appliance more likely to break, harder to use, and more costly to repair.
“The one thing a refrigerator should do is keep things cold,” Gordon-Byrne, the executive director of the Right to Repair organisation, which produces the anti-awards, wrote in the announcement.
Demonstrations at the trade show have shown that the fridge struggles to respond when there is high ambient noise, which can often occur in a busy kitchen. Organisers also pointed out the cameras on the fridge, which they deemed unnecessary.
“You don't want a camera on the front of your fridge watching you all the time,” said Elizabeth Chamberlain, the director of sustainability at iFixit, one of the groups behind the contest.
“Certainly, most people are not asking for voice-controlled, AI voice-controlled refrigerator door opening," she added.
Amazon’s Ring doorbell camera system was also named among the worst products.
The household tech was criticised over consumer surveillance linked to new AI features that were introduced across its cameras and doorbells.
Cindy Cohn, executive director of the Electronic Frontier Foundation, said the growing range of tools, including facial recognition, mobile surveillance towers, and third-party app features, risked becoming increasingly invasive.
Chamberlain said that even people without Ring devices could be affected.
"Your neighbours' cameras probably know who you are and know when you're coming home, know what you're doing, know when you’re leaving the house, when you’re going to the store, know when you’re out of town. That's creepy," she said.
Also on the Worst of Show list was Lollipop Star, an electronic lollipop that plays music as it’s licked using bone-conduction technology.
The edible device received the Environmental Impact award for its potential to create e-waste, due to its disposable batteries that only last one hour, along with single-use plastic and electronics.
“My colleague and I took it apart on the floor today. It's got two batteries in it, a speaker, a little board. It's a lot of electronics for something that's only gonna last an hour,” said Chamberlain.
The anti-awards contest was envisioned to call out technology that “undermines privacy, security, sustainability, and repairability” and has been running since at least 2023.
The judges are not affiliated with the Consumer Technology Association (CTA), which runs CES.
