Amazon Web Services (AWS) introduced its new Automated Reasoning Checks tool in December 2024, which relies on mathematical logic to check and verify the accuracy of responses generated by artificial intelligence.
Automated reasoning is a branch of symbolic artificial intelligence that traces back approximately 2,000 years to the mathematical logic work of ancient Greek philosophers such as Socrates and Plato. According to Byron Cook, AWS Vice President and Distinguished Scientist, only about 3,000 experts worldwide specialise in this field. AWS's new solution, Automated Reasoning Checks technology, was initially used primarily in cybersecurity, specifically for validating cryptographic solutions. Today, however, this procedure plays a key role in increasing the reliability of corporate AI applications by mathematically verifying system inferences. The technology is already being employed by PricewaterhouseCoopers (PwC) in areas such as pharmaceuticals and life sciences, where the accuracy of marketing materials and other communication documents is paramount due to strict regulatory requirements.
Although AWS's solution represents a significant step forward in improving AI reliability, Cook acknowledges that completely eliminating "hallucinations" generated by artificial intelligence is impossible, given our current knowledge. Therefore, in addition to Automated Reasoning Checks, he recommends that companies employ further techniques—such as retrieval-augmented generation (RAG) or model fine-tuning—to minimise the occurrence of erroneous responses. AWS's current developments are part of a broader strategy to strengthen cloud services and artificial intelligence solutions, thereby remaining competitive against rivals such as Microsoft and Google.
Sources:
1.
2.
Emilia David
3.
Saul Zimet
