I asked Chat-GPT to find me some academic papers that highlight the usefulness of combining two different machine learning techniques. I hoped that it might be able to point me to some new insights for some research we’re doing into document clustering. The results were surprising:
These look credible enough, but none of these papers exists. The researchers are fabrications and their urls don’t like to the promised documents. The answer is what Chat GPT has hallucinated into existence. That’s all very well, but there are times when you need to rely on data, when a hallucination won’t do, when its important to know that a product, a document and an entity really exists.
At Spend Network, we have government procurement data for more than 150 countries globally. We take the hard work out of collecting, organising and analysing procurement data at scale. Get in touch to find our more about our data products and services.
If you liked this post, you may like Waiting For a Tsunami of Text.