AI News

DECEMBER 10, 2024

Researchers from the Tokyo University of Science (TUS) have developed a method to enable large-scale AI models to selectively “forget” specific classes of data. Progress in AI has provided tools capable of revolutionising various domains, from healthcare to autonomous driving.

AI Modeling 297

AI Modeling Black Box AI Big Data AI 297

Unite.AI

DECEMBER 19, 2024

Thats why explainability is such a key issue. People want to know how AI systems work, why they make certain decisions, and what data they use. The more we can explain AI, the easier it is to trust and use it. Large Language Models (LLMs) are changing how we interact with AI. Thats where LLMs come in.

Large Language Models 208

Large Language Models Explainable AI Explainability Conversational AI 208

Unite.AI

MARCH 6, 2025

This success, however, has come at a cost, one that could have serious implications for the future of AI development. The Language Challenge DeepSeek R1 has introduced a novel training method which instead of explaining its reasoning in a way humans can understand, reward the models solely for providing correct answers.

Large Language Models 173

Large Language Models Explainability Black Box AI AI Researcher 173

Unite.AI

JUNE 26, 2023

At the root of AI mistakes like these is the nature of AI models themselves. Most AI today use “black box” logic, meaning no one can see how the algorithm makes decisions. Black box AI lack transparency, leading to risks like logic bias , discrimination and inaccurate results.

Black Box AI 297

Black Box AI Explainable AI Explainability AI 297

Towards AI

NOVEMBER 7, 2024

This week, we are diving into some very interesting resources on the AI ‘black box problem’, interpretability, and AI decision-making. Parallely, we also dive into Anthropic’s new framework for assessing the risk of AI models sabotaging human efforts to control and evaluate them. Enjoy the read!

Black Box AI 115

Black Box AI Machine Learning Prompt Engineer Prompt Engineering 115

Unite.AI

APRIL 9, 2024

The adoption of Artificial Intelligence (AI) has increased rapidly across domains such as healthcare, finance, and legal systems. However, this surge in AI usage has raised concerns about transparency and accountability. Composite AI is a cutting-edge approach to holistically tackling complex business problems.

Composite AI 182

Composite AI Explainability Black Box AI Explainable AI 182

Towards AI

DECEMBER 19, 2023

It is very risky to apply these black-box AI systems in real-life applications, especially in sectors like banking and healthcare. For example, a deep neural net used for a loan application scorecard might deny a customer, and we will not be able to explain why. arXiv: 2003.07132 where n is the sample size.

Neural Network 122

Neural Network AI Modeling Black Box AI Explainability 122