Investigating ChatGPT’s Business Recommendations: The Risks of Hacked and Expired Domains

Uncovering the Manipulation of Content in AI-Generated Responses

The Ethical Dilemma of AI Recommendations: Insights from James Brockbank’s Investigation into ChatGPT

In an era where artificial intelligence plays a pivotal role in how we discover products and services, a recent investigation by SEO professional James Brockbank raises critical questions about the integrity of AI-generated recommendations. His findings suggest that ChatGPT may inadvertently promote businesses using content from hacked websites and expired domains, illuminating a pressing need for vigilance and ethical considerations in the digital landscape.

The Investigation: A Dive into Manipulated Content

James Brockbank, Managing Director at Digitaloft, conducted personal testing that led to these revelations, exploring the mechanisms through which brands gain visibility in AI-generated responses. His analysis shows that certain actors are gaming the system by publishing content on compromised or repurposed domains that retain high authority signals. Despite this content being often irrelevant or outright deceptive, it can still surface in ChatGPT-generated business recommendations.

“I believe that the more we understand about why certain citations get surfaced, even if these are spammy and manipulative, the better we understand how these new platforms work.” — James Brockbank

How Manipulated Content Appears in ChatGPT Responses

Brockbank identified two prominent tactics that seem to influence ChatGPT’s business recommendations:

1. Hacked Websites

One striking example revealed that ChatGPT provided gambling recommendations traced back to legitimate websites that had been compromised. In a particularly alarming case, a California domestic violence attorney’s site was found hosting an unrelated listicle about online slots. Other notable examples include a United Nations youth coalition site and a U.S. summer camp website that had been hijacked to promote gambling-related content, employing tactics like white text on white backgrounds to evade detection.

2. Expired Domains

The second method involves acquiring expired domains with strong backlink profiles and repurposing them for unrelated content. For instance, Brockbank discovered a site with over 9,000 referring domains from reputable sources like BBC, CNN, and Bloomberg, which had been shifted from an arts charity to gambling promotion.

“There’s no question that it’s the site’s authority that’s causing it to be used as a source. The issue is that the domain changed hands and the site totally switched up.” — James Brockbank

Why This Content Is Surfacing

Brockbank’s findings suggest that ChatGPT favors domains that exhibit perceived authority and recent publication dates. The AI appears less concerned with whether the content aligns with the original purpose of the site.

“ChatGPT prefers recent sources, and the fact that these listicles aren’t topically relevant to what the domain is (or should be) about doesn’t seem to matter.”

While authentic placements in "best of" lists can help businesses gain visibility through AI, manipulating the system using hacked or repurposed domains raises ethical concerns.

“Injecting your brand or content into a hacked site or rebuilding an expired domain solely to fool a language model into citing it? That’s manipulation, and it undermines the credibility of the platform.” — James Brockbank

What This Means for Businesses and Users

Though Brockbank’s findings stem from personal observation rather than formal research, they raise significant concerns regarding the reliability of AI-generated content. The findings serve as a reminder of the importance of accountability in building brand authority through trustworthy content and ethical marketing practices.

The investigation also highlights an urgent need for AI platforms like ChatGPT to enhance their systems for detecting and filtering deceptive content. Until such improvements are made, both users and businesses should approach AI-generated recommendations with a healthy degree of skepticism.

“We’re not yet at the stage where we can trust ChatGPT recommendations without considering where it’s sourced these from.” — James Brockbank

Conclusion

As AI continues to shape how consumers discover businesses, the stakes are higher than ever. Building legitimate authority through ethical practices will be essential, not only for maintaining trust in these technologies but also for ensuring the integrity of the information that users rely on. As we navigate this evolving landscape, the responsibility falls on both companies and users to foster a digital environment that prioritizes authenticity over manipulation.

For deeper insights, check out the original report at Digitaloft.

This investigation serves as a wake-up call, urging everyone involved in the digital ecosystem to prioritize ethical practices and transparency in their strategies moving forward.