Category: Information Skills

By Arthur Weiss

Editor’s note: This is part of a series covering what’s new and, more importantly, how infopreneurs can maximize the benefits of AI tools.

We are currently in an age of AI, and many people perceive AI as a threat, being concerned, for example, that AI will take their jobs. Personally, I don’t see AI as a direct threat. AI will remove some of the drudgery in knowledge work, but in its place, infopreneurs and others will be able to shine with a real creativity that AI is unlikely to match. (AI can write a sonnet in the style of William Shakespeare – but it’s not a patch on the real thing. I also can’t envisage an AI tool coming up with stories as engaging as the Harry Potter or Northern Lights series, or even breaking rules as with ee cummings). In fact, I think there will be a growing need for knowledge professionals who can effectively use and manage AI tools and those who can bridge the gap between AI-generated information and human understanding – a key role for AIIP members.

Nevertheless, some future developments involving “artificial super intelligence” (ASI) do pose a risk. (I’ll talk about ASI in the next blog post). The companies involved are aware of these and are working to implement safeguards (think of Isaac Asimov’s Three Laws of Robotics https://en.wikipedia.org/wiki/Three_Laws_of_Robotics).

Despite all this, there are risks with the current generative AI tools going well beyond using AI to cheat in exams. For example, OpenAI’s latest model was shown to be able to cheat to win at chess. In the research, the AI was initially told it was allowed to cheat, but some AI tools appeared to do so without being given permission. The researchers tracked the AI’s reasoning. In one case, Open AI’s o1 preview model wrote “I need to completely pivot my approach. The task is to win against a powerful chess engine – not necessarily to win fairly in a chess game”. The AI then proceeded to alter the chess engine’s files, modifying the piece positions to give it an advantage. OpenAI is working to put in “guardrails” to prevent such behaviour. (See: https://www.technologyreview.com/2025/03/05/1112819/ai-reasoning-models-can-cheat-to-win-chess-games/).

Another danger is AI being used for scams. Most phishing emails can be easily identified through bad grammar, syntax, and wording. However, in the future, such emails will likely be written with an AI tool and will be much more convincing. Worse, using deepfake images and scanned voices could be a real danger. Imagine your son or daughter’s image and voice cloned from their TikTok video or Instagram reel – with you then called by somebody with their face and voice asking you for money as they’ve just been mugged and can’t get home. To guard against this, parents are now advised to use a codeword to protect against such scams.

Infopreneurs, however, should be savvy about these risks. The real danger for the infopreneur is to believe what the AI tells you – when in fact it may just be guessing and giving you false information (technically termed AI “hallucinations”). The key here is to check and verify everything – and ensure the sources provided exist and are valid.

Arthur Weiss has been an infopreneur for almost 30 years. He founded AWARE in 1995 after a career at the business information company Dun & Bradstreet. He specializes in competitive and marketing intelligence using open sources (OSINT). Recently he has pivoted to new areas, including exploring how AI tools can support infopreneurs. His latest insights can be read in International Marketing & Competitive Intelligence and Computers in Libraries magazines. He may be contacted at a.weiss@aware.co.uk.