Blog

The DeAngelis Review

Exploring Reason, Technology, and Humanity

Do Chatbots Hallucinate, Make Things Up, or Outright Lie?

This article discusses a new problem with large language models (LLMs) like Bard: AI hallucination, where the LLM generates incorrect but seemingly believable information. Experts say this is dangerous because it can spread misinformation. People can limit AI hallucination by carefully phrasing their prompts, giving the AI relevant data, and instructing it on its role. However, some argue users shouldn't have to fact-check AI outputs, and developers need to find better solutions. Large tech companies say AI needs regulation, but regulation struggles to keep pace with technology.

Water is as Important to Some Industries as Electricity

This article discusses the looming global water crisis driven by climate change, population growth, and uneven distribution. It emphasizes the risks to businesses and calls for a multi-pronged approach involving governments, industries, and NGOs to manage water more efficiently and sustainably.

Causal Artificial Intelligence and Supply Chain Performance

AI is improving supply chain forecasting by going beyond correlations to identify causal relationships. This allows for better predictions and decisions, reducing stockouts and disruptions. Causal AI uses data science to understand why things happen, not just that they happen together. This is a more powerful approach for complex systems like supply chains.

Is Artificial Intelligence Ready for Prime Time in the Supply Chain?

The pandemic exposed weaknesses in global supply chains. AI, while promising, wasn't a magic bullet. Data fragmentation between different parts of the supply chain limited AI's effectiveness. Sharing data is crucial but faces competition hurdles. AI can still help companies navigate disruptions and make better decisions.

Cognitive Computing, Decision-making, and the Supply Chain

This article explores how cognitive computing can help supply chain professionals manage information overload and complexity. Cognitive computing uses AI and machine learning to analyze data and make decisions, reducing workload for humans. It can automate routine decisions and free up professionals to make strategic choices.

Making Better Decisions with Decision Science

Although many experts insist that companies must be data-driven, they also point out that collecting data is of little value until it is analyzed for insights. The problem with most data is that it assumes the future will look like the past. That is not always true. The best analysis looks forward as well as backward. It takes current trends into account as well as past performance.