Welcome to the forefront of conversational AI as we explore the fascinating world of AI chatbots in our dedicated blog series. Discover the latest advancements, applications, and strategies that propel the evolution of chatbot technology. From enhancing customer interactions to streamlining business processes, these articles delve into the innovative ways artificial intelligence is shaping the landscape of automated conversational agents. Whether you’re a business owner, developer, or simply intrigued by the future of interactive technology, join us on this journey to unravel the transformative power and endless possibilities of AI chatbots.
A new study finds the federal government has done little to ensure its decisions are based on surveys of real people — not bots.
“Bot infiltration” is a growing problem in public opinion polling, and new research finds the federal government isn’t keeping pace with the scale of the threat.
Subscribe now to read the latest news in your city and across Canada.
Subscribe now to read the latest news in your city and across Canada.
Create an account or sign in to continue with your reading experience.
Create an account or sign in to continue with your reading experience.
The federal government commissions hundreds of surveys a year seeking insight into what Canadians think about a wide range of topics.
Increasingly, researchers are finding such online public opinion surveys are vulnerable to exploitation by artificial intelligence — bots that can mimic human responders and potentially sway the results.
Safeguards baked into surveys commissioned by the government, meanwhile, aren’t keeping pace, according to a new research paper.
“When it’s the government doing a survey, it’s not just to understand the world better, it’s also to make decisions,” said Étienne Charbonneau, an author on the study and professor at the National School of Public Administration in Montreal.
“If the decisions are based on the point of view of bots, then they might take decisions that are not popular with actual Canadians.”
The consequences are widespread, potentially undermining the evidence behind government decisions on everything from immigration to taxation.
On “sensitive topics,” Charbonneau said, interest groups could wield bots to try and sway the government’s perspective.
Other researchers and pollsters have highlighted examples where a hostile state may try to influence foreign policy.
In cases where a survey offers a financial incentive, fraudsters may simply be trying to make a quick buck.
The peer-reviewed study, published in the International Public Management Journal, looks at efforts taken by polling firms working on behalf of government departments and agencies to “keep bots at bay.”
It looked at more than 700 public opinion research reports worth a combined $120 million that were commissioned by the federal government between 2020 and 2025.
It found that since the launch of ChatGPT-3, considered a leap forward in the field of artificial intelligence, anti-bot safeguards “have made little progress and remain rare.”
While the use of some technical controls, such as device tracking, increased, more advanced methods, such as email verification, did not.
But even the more advanced bot-detection techniques, Charbonneau said, are quickly becoming out of date.
“It’s really hard to know what still works now, today, as we talk,” he said in an interview. “It’s hard to say what safeguards are still effective.”
Frank Graves, the president and founder of polling firm EKOS Research, called the study “well-executed” and said its findings point to an application of AI that he views as “kind of disturbing.”
Recent research, he said, has shown that synthetic AI can be fashioned in a way that “renders most current opt-in panels, some more so than others, defenceless.”
“This is a concern,” he said.
Although AI is perhaps more difficult to deal with, Graves said it is just one of “a plethora of different things” bringing the overall quality of polling down.
Curtis Brown, a vice-president at Ottawa-based polling firm Abacus Data, said his company often works with online survey panels and is constantly thinking about the issue of bots polluting survey data.
Abacus, he said, employs quality control checks and flags where AI is being used, often working with a platform called PureSpectrum that assigns a quality control score to responses from online panels.
“They have really rigorous and transparent checks in terms of how they manage quality control,” he said, adding that human experts ensure the firm takes in quality material.
But Charbonneau said AI advances are outpacing safeguards.
Bots can mimic human respondents by matching reading times to a given education level, generating human-like mouse movements and inserting believable typos and corrections.
As Charbonneau sees it, the solution is to conduct surveys as much as possible over the phone or hold in-person focus groups. The government could also bake higher standards into contracts when it puts out a call for bids.
Michèlle LaRose, a spokesperson for Public Services and Procurement Canada, said in an email that the government is “incorporating additional guardrails” against the use of AI.
“This includes updates to our research guidance, strengthened statements of work and research proposals, increased data quality measures, transparency about the use of AI in GC research activities and ongoing engagement with standing offer suppliers and the Canadian Research Insights Council,” she said.
Graves said any survey used for government policy should be able to establish whether the respondent is a real person and where the data came from.
Charbonneau recognizes the tradeoffs. Phone surveys are costlier, he said, and have a harder time reaching young people. But for him, the drawbacks are worth it.
“If a topic is important enough to spend that money, maybe it’s worth spending more to have real Canadians answer these questions,” he said.
Don’t let algorithms hide the follow-up. If you are tracking this story, make the Ottawa Citizen a preferred source so you get our live updates the next time you search.
Postmedia is committed to maintaining a lively but civil forum for discussion. Please keep comments relevant and respectful. Comments may take up to an hour to appear on the site. You will receive an email if there is a reply to your comment, an update to a thread you follow or if a user you follow comments. Visit our Community Guidelines for more information.
365 Bloor Street East, Toronto, Ontario, M4W 3L4
© 2026 Ottawa Citizen, a division of Postmedia Network Inc. All rights reserved. Unauthorized distribution, transmission or republication strictly prohibited.
This website uses cookies to personalize your content (including ads), and allows us to analyze our traffic. Read more about cookies here. By continuing to use our site, you agree to our Terms of Use and Privacy Policy.