Black Box
A system whose internal workings are unknown or ignored, analyzed only by its inputs and outputs.
As defined in et_941
Other definitions
A system whose internal decision-making processes are not transparent or interpretable to users or regulators—in this context, applied to large language models whose outputs cannot easily be explained.
Jaron Lanier Looks into AI's Future | AI IRLA system whose internal workings are opaque to users, so only inputs and outputs are visible, which can make it seem magical or valuable.
Why Venture is Hard
Where it comes up · 26
- Tracing the thoughts of a large language model4 claims
- Scarcity, Abundance and Bubbles2 claims
- Cliff Asness: A Brief and Biased Survey of Quantitative Investing2 claims
- 319 - The Digital Multiverse2 claims
- 46 | Kate Darling on Our Connections with Robots2 claims
- Navigating Our Revolutionary Age: A Conversation with Fareed Zakaria and Eric Schmidt1 claim
- Humanity’s Final Goal | Isaac Asimov’s The Last Question1 claim
- Creating and Selling Superpowers1 claim
and 18 more
Most important claims · 10 of 33Search all claims for “Black Box” →
- Unlike computers, which humans invented from scratch, living systems were not designed by us, so we engineer them as black boxes, observing and gradually learning to exploit them without fully understanding their logic.Scarcity, Abundance and Bubbles
- Today's deep learning systems are powerful but inscrutable black boxes that do impressive things without our understanding why, like children whose neural networks learn to speak English in ways we cannot explain; Tegmark's MIT research focuses on AI transparency—'intelligible intelligence'—transforming neural networks into more understandable systems that perform just as well, which is needed to trust everything from judicial rulings to infrastructure-controlling machines.94 - Frontiers Of Intelligence
- Most of the 10,000 tech experiments are not designed to uncover fundamental generalizable truths about the world; they are specific to a firm's interface, often comparing two black-box algorithms where you never learn why one worked better, so even if published they would not yield generalizable parameters like risk aversion or class-size effects that economists seek.Susan Athey on Machine Learning, Big Data, and Causation
- Opening the black box of AI systems without interpretability tools is unhelpful, analogous to a neuroscientist investigating the brain without proper instrumentation—tools are needed to understand what is actually happening inside the system.Tracing the thoughts of a large language model
- We must understand how AI works rather than accepting it as a black box, because AI systems make decisions without transparency about their mechanisms, and trusting a system that tells you 'you have cancer' but can't explain why requires understanding the decision mechanism—otherwise we return to a pre-enlightenment world where authority is unquestionable.Navigating Our Revolutionary Age: A Conversation with Fareed Zakaria and Eric Schmidt
- Mitchell's core questions about future AI are: Can we train systems to better understand our world, our values, and our intentions? Can we develop scientific tools to understand how these systems actually work internally? These remain unanswered and are the focus of her researchThe Future of Artificial Intelligence
- Current AI driving systems can be trivially fooled in ways no human would be — reading a Burger King sign as a stop signal, mistaking pictures of bicycles on a van for real bicycles, or being misled by a hand-drawn fake sign — because machine learning systems are black boxes that mimic intelligence rather than possessing common sense.109 | Jason Torchinsky on Our Self-Driving Future
- Deep learning systems are largely black boxes, raising three concerns: practical debugging is hard when failures can't be explained, legal and ethical acceptability suffers in high-stakes domains (medical diagnosis, parole, credit) where explanations and bias detection are required, and—most importantly for safety—opacity removes our ability to know what increasingly capable systems are doing, eroding control.53 - The Dawn Of Artificial Intelligence
- AI systems are often described as black boxes where words go in and words come out, but the internal reasoning process is opaque because AIs are trained rather than programmed, meaning they learn their own strategies to solve problems during training rather than following explicit rules.Tracing the thoughts of a large language model
- We currently have a decision-making process without an auditing process for it; we need to build tools to audit algorithms—companies should audit their own algorithms internally to avoid illegal or unfair outcomes, and regulators should recognize these algorithms are running wild and need to be understood.Cathy O'Neil on Weapons of Math Destruction