In October 2025, Deloitte was thrown into its biggest scandal in recent years.
The consultancy prepared a welfare report for Australia’s Department of Employment and Workplace Relations, which was rife with AI-generated fabrications. The report, commissioned for nearly US$300,000 (about AU$440,000), erroneously included fake citations and a fabricated court quote that surfaced only after an external scrutiny.
Deloitte acknowledged the failure, stopping short of a full apology. “We take responsibility for the fact that appropriate review and oversight processes were not followed on this occasion,” the firm said, adding that its people “remain accountable for the accuracy and integrity of our work, regardless of the tools used.”
A partial refund followed, along with a corrected report.
While the incident is 2025 news and Deloitte has since moved to stabilise its reputation, the episode has become a reference point for brands and institutions that face wider anxiety on what happens when AI speeds ahead of oversight, and credibility becomes the weakest link.
For HS Chung, the newly minted APAC chief executive at Burson, AI blunders are not just about the failure of technology, but also about the challenges of managing reputation in an environment where these tools accelerate both misinformation and scrutiny.
Chung stepped into the APAC CEO role in November 2025, expanding her remit from North Asia Pacific to oversee Burson’s full regional footprint from Australia and New Zealand through to India, Southeast Asia, Greater China, Japan and Korea.
That breadth forms her view on why credibility is a non-negotiable in today's environment.
“The value of having credibility and reputation, whether you’re a corporation or a brand, has never been as important as it is today,” she says in her first post-CEO interview with Campaign Asia-Pacific. “That reputation lever essentially rests on credibility. Consumers now, more than ever, need to be able to trust the information they receive."
For Chung, the backdrop to AI-related controversies is a radically changed information ecosystem. “Everything is becoming fragmented," she added. “Identifying what is real and factual is becoming more and more complicated because there is just so much information.”
Reputation and visibility in the GEO era
“In the past, when everything revolved around differentiation, pushing brands and products, reputation was often treated as something ‘nice to have,’ maybe managed at the corporate level,” she explained. “But now, the reputation of any brand or product has become one of the most important factors because it establishes credibility more than anything else.”
This applies as much to AI mistakes as to traditional crises. In an AI blunder, communicators need to quickly diagnose which of those levers are most at risk—governance, innovation, or product quality—and respond in a way that restores credibility on those dimensions, rather than simply issuing technical clarifications.
One reason incidents like Deloitte’s flared so quickly is the shift in how people discover and evaluate information. Chung said many markets were “heavily fixated on SEO‑focused search and SEO‑based marketing,” but that pattern is changing.
“Search revenue and user activity are declining rapidly,” she noted. “At the same time, we’re seeing the rise of GEO engines, with AI platforms like ChatGPT, Gemini, Claude, and Perplexity. The usage rate of these platforms is increasing dramatically. Traditional search is declining while generative platforms, where users prompt, ask questions, and receive direct answers, are increasing.”
In the GEO era, when an issue breaks, generative platforms pull from regulators, major publishers and authoritative brand statements to form instant answers. A government inquiry in Australia, a policy response in Singapore or a CEO statement picked up by regional media can quickly become the default narrative surfaced by AI. Once embedded, those signals are hard to reverse. This raises the stakes for early, credible intervention over reactive clean-ups.
“One of the biggest differences between the SEO era and what we’re calling the GEO era is that GEO visibility cannot be bought, “ she observed. "You can’t pay for it, you can’t buy it through advertising or media spending. The majority of visibility in the GEO space comes from earned and owned media, structured, credible, and strategically managed content.”
Asked what a Crisis 101 response now looks like, Chung returned to fundamentals: “From a crisis management perspective, credibility is everything."
“An official company message communicated by a credible CEO or spokesperson is critical. AI platforms treat this kind of authoritative information as credible and are more likely to surface it. So consistency, authenticity, and clarity in messaging have never been more important," she added.
In addition, brands should be aware of how they show up in AI environments. This becomes critical when AI‑related mistakes surface. With AI systems already associating a brand with high‑quality, transparent, well‑governed content, a single failure is more likely to be seen in its algorithm as an exception.
Chung's recommendation is to assess visibility within AI platforms: "Even companies that think they’re performing well often find that their visibility across these platforms differs completely from traditional search results.”
AI as an enabler, not a shield
Chung is clear that AI can strengthen crisis insight, even in the event of AI blunders, but only if communicators stay in charge.
She noted: “With AI, we can gather sufficient and accurate data faster, at a scale that wasn’t possible before. The scale, precision, and accuracy of data we can now access, combined with methodology and insight, allow us to provide business‑critical insights. This goes far beyond discussing media coverage.”
For example, Burson's Reputation Capital, an AI-driven platform and consulting framework, helps clients understand the relationship between reputation drivers and business outcomes like stock price, sales, and purchase intent. Chung argued that tools like these, combined with human intelligence, enable faster insights with more precise audience targeting that helps clients drive business impact.
At the same time, Chung warned against over‑reliance on tools when reputations are on the line: “As now so many tools are available, some people think everything can be automated, that tools can replace human effort or decision‑making. This overreliance gives a false sense of efficiency.”
“No tool can replace the nuanced thinking, emotion, or creativity that humans bring,” she added. “AI tools are enablers; they make your job faster and more accurate, but they don’t replace human insight. We, as humans, must remain in the driver’s seat. We have to lead the tools, not be led by them.”
For marketers and communicators, her advice is blunt: define the brand objective first, treat AI as an enabler, and anchor every response in credibility. “Going into 2026, brands need to cut through the noise and first define what their brand objective is,” she said, adding: “The right approach is to see AI as an enabler that helps you do your job better, not a replacement for your strategic thinking.”
