Definition
Within artificial intelligence (AI), explainable AI (XAI), generally overlapping with interpretable AI or explainable machine learning (XML), is a field of research that explores methods that provide humans with the ability of intellectual oversight over AI algorithms. The main focus is on the reasoning behind the decisions or predictions made by the AI algorithms, to make them more understandable and transparent. This addresses users' requirement to assess safety and scrutinize the automated decision making in applications. XAI counters the "black box" tendency of machine learning, where even the AI's designers cannot explain why it arrived at a specific decision.
Related concepts
15.aiACM Conference on Fairness, Accountability, and TransparencyAI-assisted software developmentAI agentAI alignmentAI anthropomorphismAI boomAI bubbleAI data centerAI effectAI literacyAI nationalismAI safetyAI slopAI takeoverAI veganismAI winterAcademic PressAccumulated local effectsAction selectionActivation functionAdobe FireflyAdversarial machine learningAgency (sociology)Agent2AgentAidan GomezAlan TuringAlexNetAlex Graves (computer scientist)Alex KrizhevskyAlgorithmAlgorithmic transparencyAllen NewellAlphaFoldAlphaGoAlphaZeroAndrej KarpathyAndrew NgApplications of artificial intelligenceAriel D. ProcacciaArtificial Intelligence ActArtificial Intelligence Cold WarArtificial consciousnessArtificial general intelligenceArtificial human companionArtificial intelligenceArtificial intelligence and electionsArtificial intelligence arms raceArtificial intelligence artArtificial intelligence controversiesArtificial intelligence in architectureArtificial intelligence in educationArtificial intelligence in fictionArtificial intelligence in governmentArtificial intelligence in healthcareArtificial intelligence in industryArtificial intelligence in mental healthArtificial intelligence in video gamesArtificial intelligence systems integrationArtificial intelligence visual artArtificial neural networksArtificial neuronArtificial superintelligenceAshish VaswaniAttention (machine learning)Aurora (text-to-image model)AutoGPTAutoencoderAutomated Pain RecognitionAutomated decision-makingAutomated planning and schedulingAutomated reasoningAutomated theorem provingAutoregressive modelAxiomBERT (language model)BLOOM (language model)BackpropagationBacteremiaBatch normalizationBayesian networkBernard WidrowBias–variance tradeoffBitter lessonBlack boxBloomberg BusinessweekBorda votingChatbot psychosisChinchilla (language model)Chinese roomChristopher D. ManningCircuit (neural network)Claude (language model)Claude ShannonCliff ShawClinical decision support systemCluster analysisCompetition in artificial intelligenceComputer visionConjugate gradient methodConvolutionConvolutional neural networkDALL-EDARPADBRXDaniel Day-LewisDaniel Kokotajlo (researcher)Data augmentationDavid Silver (computer scientist)Decision treeDeepDreamDeepSeek (chatbot)Deep learningDeep learning speech synthesisDeepfakeDeepfake pornographyDefense industryDemis HassabisDictionary learningDifferentiable neural computerDiffusion processDouble descentDream Machine (text-to-video model)Echo state networkElevenLabsEnvironmental impact of artificial intelligenceEthics of artificial intelligenceEvidence-based medicineEvolutionary algorithmExistential risk from artificial general intelligenceExpert systemFacial recognition systemFei-Fei LiFinanceFlux (text-to-image model)Foundation modelFrank RosenblattFrançois CholletFriendly artificial intelligenceFrontier modelGPT ImageGated recurrent unitGating mechanismGemini (chatbot)Gemini (language model)Gemma (language model)General Data Protection RegulationGeneral game playingGenerative adversarial networkGenerative artificial intelligenceGenerative audioGenerative engine optimizationGenerative pre-trained transformerGenie (world model)Geoffrey HintonGlass boxGloVeGlossary of artificial intelligenceGoogle Gemini image generation controversyGradient descentGraph neural networkGrok (chatbot)Grok sexual deepfake scandalGrokking (machine learning)Hallucination (artificial intelligence)Handwriting recognitionHerbert A. SimonHighway networkHistory of artificial intelligenceHuawei PanGuHuman-in-the-loopHuman image synthesisHumanity's Last ExamHuman–AI interactionHybrid intelligent systemHyperparameter (machine learning)IBM GraniteIBM WatsonIBM WatsonxIan GoodfellowIdeogram (text-to-image model)Ilya SutskeverImage recognitionImagen (text-to-image model)Imitation learningIntelligent agentIntelligent tutoring systemsInternational Joint Conference on Artificial IntelligenceInternational Organization for StandardizationIt's the Most Terrible Time of the YearJames GoodnightJan LeikeJohn HopfieldJohn McCarthy (computer scientist)John SchulmanJohn von NeumannJoseph WeizenbaumJürgen SchmidhuberKlaus-Robert MüllerKling AIKnowledgeKnowledge extractionKnowledge representation and reasoningKunihiko FukushimaLaMDALanguage modelLarge language modelLatent diffusion modelLawLethal autonomous weaponLinear modelList of artificial intelligence companiesList of artificial intelligence projectsLlama (language model)Loi pour une République numériqueLong short-term memoryLoss functions for classificationLossy compression artefactLotfi A. ZadehMachine learningMachine learning in bioinformaticsMachine learning in earth sciencesMachine learning in physicsMachine translationMamba (deep learning architecture)Marvin MinskyMechanistic interpretabilityMedicineMethod of equal sharesMidjourneyMiniMax (company)Model Context ProtocolMuZeroMultilayer perceptronMultitask learningMusic and artificial intelligenceMustafa SuleymanMycinNathaniel Rochester (computer scientist)Natural language processingNeural Turing machineNeural machine translationNeural network (machine learning)Noam ShazeerNormalization (machine learning)Oasis (Minecraft clone)Oliver SelfridgeOpen-source artificial intelligenceOpenAI FiveOptical character recognitionOriol VinyalsOverfittingPaLMParameterParticipatory budgetingPaul WerbosPhilosophy of artificial intelligencePolicy gradient methodPrecautionary principleProgress in artificial intelligenceProject DebaterPrompt engineeringQ-learningQuasi-Newton methodQuoc V. LeQwenReasoning modelRecraftRectifier (neural networks)Recurrent neural networkRecursive self-improvementReflection (artificial intelligence)Regression analysisRegularization (mathematics)Regulation of artificial intelligenceRegulation of artificial intelligence in the United StatesReinforcement learningReinforcement learning from human feedbackRemoval of Sam Altman from OpenAIResidual neural networkRetrieval-augmented generationReverse engineeringRiffusionRight to explanationRobot controlRoboticsRule-based systemRunway (company)SPICESaliency mapSeedance 2.0Self-driving carSelf-supervised learningSentiment analysisSeppo LinnainmaaSeq2seqSeymour PapertShapley valueShun'ichi AmariSigmoid functionSocial choice theorySoftmax functionSora (text-to-video model)Speech recognitionStable DiffusionStatement on AI RiskState–action–reward–state–actionStatistical classificationStatistical inferenceStephen GrossbergStochastic gradient descentSu-In LeeSuno (platform)Supervised learningSupervised machine learningSymbolic artificial intelligenceSymbolic regressionT5 (language model)Takeo KanadeTay (chatbot)Taylor Swift deepfake pornography controversyText-to-image modelText-to-video modelThe Alignment ProblemThéâtre D'opéra SpatialTimeline of artificial intelligenceTraining, validation, and test data setsTransformer (deep learning)Truth-maintenance systemsTuring testUdioUncanny valleyUser experienceVariational autoencoderVeo (text-to-video model)Vibe codingVirtual politicianVision transformerVoiceverse NFT plagiarism scandalWalter PittsWarren Sturgis McCullochWaveNetWeak artificial intelligenceWeight initializationWhisper (speech recognition system)White-box testingWord2vecWord embeddingWorkplace impact of artificial intelligenceWorld model (artificial intelligence)Xiaomi MiMoYann LeCunYoshua Bengio
20 concepts already in your glossary