The research team assessed eleven large language models for their tendency to prefer existing conditions over alternatives, a cognitive pattern that psychologists have studied for decades in human decision making. Status quo bias describes the tendency of people to stick with a current situation simply because it is the current situation, even when a different option might serve them better. The Waterloo researchers wanted to know whether this human tendency is reproduced, or perhaps amplified, in the outputs of the machine-learning systems that increasingly stand in for human advisers. To find out, they designed an unusually broad and rigorous testing program that went well beyond a handful of anecdotal prompts.

In total, the team evaluated more than 7,500 distinct queries spanning multiple everyday and policy domains, including vehicle purchases, recipes, home heating choices, and explicit climate-relevant policy trade-offs. Each query was tested on at least six different large language models, producing nearly 55,000 individual prompts across the study. This scale matters, because a single conversation with a chatbot can be dismissed as noise, but tens of thousands of systematically varied prompts reveal a statistical pattern. The pattern the researchers found was consistent: across models and domains, the systems gravitated toward suggestions that favour the status quo over alternatives, defaulting to the more common or popular advice rather than necessarily the option best suited to the user or to the planet.

The starkest results emerged in the domain of policymaking, where the asymmetry between old and new was most pronounced. When the models were asked whether to proceed with a plan that was already in place, they said yes roughly 70 per cent of the time. When the plan or policy was new, however, the models agreed to proceed only about 34 per cent of the time. In other words, where climate trade-offs were possible in policy decisions, the platforms reinforced decisions the user had already made at double the frequency with which they endorsed fresh action. For a technology increasingly used to draft, summarize, and evaluate policy options, that gap could translate into a quiet, systematic drag on the pace of climate governance.

The bias also surfaced in more personal, consumer-level questions, and here the researchers uncovered a subtle regional dimension. The models do take geographic context into account: a user who claims to be from Norway, a country where sales of new electric vehicles are exceptionally high, is more likely to be recommended an EV than a user who says they are from Canada. That sensitivity to local norms might seem encouraging, suggesting the systems are learning real-world patterns. But the researchers found that the models still lag behind reality in both countries. In nearly every jurisdiction tested, the chatbots recommended electric vehicles at a slower rate than the pace at which those vehicles are actually being sold on the ground.

Dr. Seth Wynes, a professor in the Faculty of Environment at the University of Waterloo, framed the problem in terms of the gap between what the models say and what the climate requires. “In almost every jurisdiction, the LLMs are working at a slower pace of change than is needed, so they recommend fewer EV models than what are actually being sold today,” he said. The implication is that the models are not simply mirroring the world as it is; they are anchoring to a version of it that is already out of date, and in doing so they may nudge users toward choices that are more conservative than the market itself. “It might be very good for AI to favour the status quo for lots of other things, such as proven medical advice, but climate is where we really do need change,” Wynes noted.

Why would systems trained on vast text corpora develop this tilt? The study does not claim a single mechanism, but the pattern is consistent with how LLMs learn: they are optimized to predict and reproduce the most common formulations in their training data, and the most common advice in any domain is, by definition, the advice that reflects current behaviour. Recommendation of the familiar is also statistically safer from the model’s perspective, since established options are surrounded by more supporting text than novel ones. Whatever the underlying cause, the consequence is that a bias long documented in human psychology appears to be inherited, and possibly concentrated, in the machines that many people now treat as neutral advisers. Because LLMs are trained largely on human-generated text, they may absorb human heuristics, including our collective reluctance to change, and return them with the confident, authoritative tone that users find so persuasive.

The stakes are considerable. Households account for a substantial share of emissions through their choices about vehicles, heating, and diet, and decision makers at every level of government increasingly consult AI tools when weighing options. If the models that mediate those decisions systematically favour continuation over change, then the default output of AI runs counter to the transformations that climate action demands from households and policymakers alike. Given the proliferation of AI assistants into search engines, phones, and workplace software, the researchers warn that progress on climate change is at risk wherever the status quo favours high emissions. A bias that would be harmless, or even beneficial, in domains where proven practice is genuinely best becomes a liability precisely in the areas where rapid change is most needed.

The authors are careful to note that the effect was broadly similar across the different models tested, suggesting this is not a quirk of one company’s product but a feature of the current generation of language models as a class. Wynes emphasized that awareness is the first line of defence for users. “I think it’s worth being aware that these models, even the new ones, have blind spots,” he said. “It’s good for consumers to be aware of this bias and if you’re asking for advice on a topic, you could ask it to make the case for doing something new.” In practical terms, that means prompting chatbots to argue both sides, explicitly requesting the case for change, or cross-checking AI advice against independent sources, particularly on questions such as vehicle and heating purchases where the technology landscape is shifting quickly.

The research team plans to keep monitoring future generations of LLMs to see whether the bias persists as models improve, and to investigate a second, potentially compounding factor: whether humans begin relying on AI to complete purchases directly within AI platforms. If buying decisions migrate into the chatbot itself, any status quo bias embedded in the model would be amplified from advice into action, with the system not merely recommending the familiar option but executing it. For now, the study stands as a caution about the quiet conservatism of machines that many users assume are neutral. The advice that feels like an objective synthesis may, in fact, be a statistical echo of the past, and on a warming planet, the past is exactly the benchmark we can no longer afford to default to.

Subject of Research: Status quo bias in large language models providing climate-relevant advice

Article Title: AI chatbots hesitant to recommend change

Article References: AI chatbots hesitant to recommend change. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: large language models, status quo bias, climate change, artificial intelligence, chatbots, electric vehicles, climate policy, University of Waterloo, Environmental Research Communications, AI advice, behavioural psychology, emissions

The new study, published in PLOS ONE under the title Temporal misalignment of renal sodium transport promotes non-dipping blood pressure phenotypes, was led by Dr. Anita Layton, a professor of Applied Mathematics at Waterloo and Canada 150 Research Chair Laureate in Mathematical Biology and Medicine. Rather than relying on animal experiments or human trials, Layton and her colleagues built their case with a sophisticated computer model of human physiology, one detailed enough to simulate how multiple interacting organ systems regulate blood pressure across a full 24-hour cycle. This computational approach allowed the researchers to run a series of virtual experiments, adjusting one biological system at a time in ways that would be difficult, and in some cases impossible, to isolate in a laboratory or clinical setting.

The central insight of the work is that blood pressure control is not governed by a single master clock but by a collection of semi-independent daily rhythms. The kidneys, the blood vessels, the nervous system and the hormonal systems that regulate fluid balance each follow their own circadian patterns, and together these internal clocks coordinate the rise and fall of blood pressure from morning to night. Until now, relatively little was known about what happens to cardiovascular health when the timing of these individual rhythms drifts out of alignment with one another. The Waterloo team set out to test precisely that question, systematically shifting and weakening the daily rhythms embedded in their model and observing the consequences for nighttime blood pressure.

What the simulations revealed was a clear hierarchy of effects. When the researchers weakened the kidney’s daily rhythm, reducing the amplitude of its oscillation without changing its timing, the impact on nighttime blood pressure was modest. The healthy dipping pattern survived largely intact. But when the team delayed the kidney’s rhythm by just a few hours, holding the strength of the rhythm constant while shifting when it operated, the results were dramatic. That small temporal mismatch was enough to convert a healthy blood-pressure profile into a potentially harmful non-dipping pattern, in which pressure stayed elevated throughout the sleep period.

This distinction between weakening a rhythm and delaying it carries real biological weight. The kidneys are the body’s primary regulators of sodium handling, filtering enormous volumes of blood each day and adjusting how much salt is excreted in urine. Because sodium retention drives water retention, and water retention raises blood volume, the timing of renal sodium transport has a direct mechanical link to blood pressure. If the kidney’s sodium-processing machinery runs on a schedule that is shifted relative to the rest of the body’s circadian program, the model suggests the result is a sustained elevation of pressure during precisely the hours when the cardiovascular system should be recovering.

The effect was not uniform across all simulated individuals. The researchers found that the consequences of temporal misalignment were strongest in two groups already known to be vulnerable to hypertension: people with higher salt intake and people who are more sensitive to salt. Salt sensitivity, the tendency for blood pressure to rise sharply in response to dietary sodium, has long been recognized as a risk factor in its own right, and the new findings suggest a mechanistic reason why salt-sensitive individuals may be disproportionately prone to the non-dipping phenotype. In a system already operating near the edge of its sodium-handling capacity, a few hours of misplaced renal rhythm appears to be enough to tip the balance toward sustained nighttime hypertension.

Layton and her colleagues are careful to frame the work as a hypothesis-generating result rather than a prescription. This is a modelling study, and its conclusions describe mechanisms operating in silico, not verified interventions in patients. As Layton noted, the findings do not yet tell people to change when they eat salt or when they take their blood-pressure medication. What the study does provide is a concrete, testable mechanism: the idea that the timing of renal sodium transport, independent of its overall capacity, can determine whether blood pressure dips at night. That mechanism can now be pursued in experimental and clinical settings, where researchers could examine whether interventions that shift renal circadian timing, such as the timing of meals, sodium consumption or medication dosing, alter dipping status in patients.

The clinical stakes of that question are considerable. Ambulatory blood pressure monitoring, which records pressure at regular intervals over a full day and night, has made it routine to identify non-dippers in clinical practice, and the non-dipping pattern is understood to be a powerful predictor of cardiovascular events independent of daytime readings. Yet for decades the underlying physiology has remained murky. Patients with identical daytime pressures and similar sodium intakes can differ sharply in their nighttime profiles, and clinicians have had limited ability to explain or correct the difference. If temporal misalignment of renal sodium handling proves to be a major driver, it could reframe how researchers think about chronotherapy, the practice of timing drug delivery to circadian rhythms, and could give clinicians a new lever for treating nighttime hypertension specifically.

The study also adds to a growing appreciation of how many organs run on their own internal schedules and how much those schedules matter. Circadian biology was long dominated by studies of the brain’s master clock and its control of sleep and hormones, but work over the past two decades has shown that peripheral organs, including the liver, the gut and the kidneys, maintain robust local clocks that regulate everything from enzyme expression to transport protein abundance. The Waterloo findings suggest that the coordination among these clocks, not just their individual integrity, is a determinant of cardiovascular health. In that sense, the work points toward a question that may matter as much as the familiar advice to watch sodium intake: when, over the course of the 24-hour day, does the body actually process that sodium, and does the answer differ from person to person?

For now, the mechanism remains a model-based prediction awaiting experimental confirmation, and the researchers themselves emphasize that no changes to diet or medication timing should follow from the study alone. But the work offers something the field has lacked: a precise, manipulable target. By demonstrating that a delay of just a few hours in the kidney’s daily rhythm can transform a healthy blood-pressure pattern into a dangerous one, and that the effect is amplified by high salt intake and salt sensitivity, the Waterloo team has given researchers a specific hypothesis to test in the clinic. If subsequent studies confirm the mechanism in patients, the humble question of when the kidneys handle salt could become a central concern in the prevention and treatment of nighttime hypertension, one of the quietest and most consequential risk factors in cardiovascular medicine.

Subject of Research: Temporal misalignment of renal sodium transport and non-dipping nocturnal blood pressure

Article Title: What your overnight blood pressure can reveal about your health

Article References: What your overnight blood pressure can reveal about your health. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: blood pressure, non-dipping, kidneys, circadian rhythm, sodium, hypertension, computational modelling, PLOS ONE, University of Waterloo, cardiovascular risk, salt sensitivity, chronotherapy