Artificial intelligence applications

See also: Terms, Models and Guides Artificial intelligence / AI apps

Artificial intelligence applications are the real-world deployments of AI techniques across industry, science, and consumer software: reading chest X-rays, scoring card transactions for fraud, driving robotaxis, designing proteins, answering customer questions, and writing code. By 2025 these uses had moved from pilots to infrastructure: Stanford's AI Index reported that 78% of organizations used AI in at least one business function in 2024, up from 55% a year earlier, and McKinsey's State of AI survey put organizational AI use at 88% in 2025.^[19][20] The same underlying methods (deep learning, transformer-based large language models, diffusion models, reinforcement learning, and computer vision) are reused across sectors that look nothing alike. This page organizes the major categories of AI apps both by sector and by modality (text, image, audio, video, code), and includes representative products in each area.

What are the main applications of artificial intelligence?

The following table summarizes the largest AI deployment areas, the kinds of tasks AI handles in each, and a few representative systems. Sector-specific detail and citations are in the sections below.

Scroll sideways for more →

The scale of investment behind these deployments grew just as fast as adoption. Stanford's AI Index put total corporate AI investment at $252.3 billion in 2024, with private investment in generative AI alone reaching $33.9 billion, more than 8.5 times the 2022 level and over 20% of all AI-related private investment.^[19]

What is AI used for in healthcare?

Healthcare is one of the most regulated AI markets and also one of the fastest growing. As of December 2025 the U.S. Food and Drug Administration's public list contained 1,451 AI-enabled medical devices authorized for marketing since it began tracking them, up from over 1,250 in mid-2025 and roughly 950 in August 2024. Radiology dominates the list, accounting for about 76% of all authorizations (1,104 of 1,451 devices), and remained roughly three quarters of new clearances throughout 2025.^[1][2][21]

Typical clinical applications include computer-aided detection of lung nodules, intracranial hemorrhage, breast cancer, and bone fractures on X-ray, CT, MRI, and ultrasound. Aidoc's CARE1 became the first foundation-model-powered clinical AI to receive FDA clearance on February 18, 2025, for rib fracture triage.^[1][22] Philips received FDA 510(k) clearance the same year for SmartSpeed Precise, a deep learning MRI reconstruction pipeline that reduces scan time while preserving image detail.^[3]

Beyond imaging, AI is used in drug discovery, clinical documentation, and decision support. AlphaFold 2 predicts the 3D structure of a protein in hours instead of months, and the AlphaFold Protein Structure Database now covers more than 214 million sequences. AlphaFold 3, released in 2024, predicts not only protein structures but also interactions with ligands, nucleic acids, and antibodies, opening up structure-based drug design for targets that have no experimental structure. Google DeepMind described AlphaFold 3 as a model that "predicts the structure and interactions of all of life's molecules."^[5] The work was recognized with the 2024 Nobel Prize in Chemistry.^[4][5] (See AI in healthcare for the full overview.)

Scroll sideways for more →

How is AI used in finance?

Financial services adopted machine learning early; the use of AI in banking is often dated to 1987 when Security Pacific National Bank launched an AI fraud prevention task force. Today more than 85% of financial firms actively apply AI somewhere in their operations, and AI spending in financial services is projected to reach roughly $97 billion by 2027.^[6]

Key domains include fraud detection, anti-money laundering (AML), credit scoring, algorithmic trading, and customer-facing assistants. Algorithmic and high-frequency trading account for over 60% of U.S. equity trades, with machine learning models analyzing tick data and news feeds in milliseconds. Recent fraud detection systems achieve precision above 95% while reducing financial fraud losses by as much as 50%.^[6] Banks that use Visa Advanced Authorization have reported fraud losses falling by about 35%, while JPMorgan Chase has reported a roughly 20% reduction in false positives from machine learning models, cutting friction for legitimate customers.^[23]

Large banks have also begun rolling out internal generative AI tools. JPMorgan Chase's LLM Suite reached more than 200,000 employees within roughly eight months of launch, one of the largest corporate AI deployments on Wall Street, and was used in Q1 2025 for credit decisions, fraud prevention, liquidity management, and document review, with the bank reporting nearly $1.5 billion in cost savings.^[6][23] (See AI in finance for more detail.)

Scroll sideways for more →

How is AI used in transportation and logistics?

Autonomous driving is the marquee AI application in transportation. Waymo crossed 450,000 paid robotaxi rides per week in December 2025, roughly double its April 2025 figure of 250,000, and reached about 500,000 weekly rides by March 2026 with a stated goal of 1 million for the year. The company operates a fleet of around 2,500 fully driverless vehicles across more than ten U.S. cities while planning service or testing in additional markets.^[7][24] In late 2025 Waymo pulled human safety drivers from its vehicles in Dallas and Houston, restricting them to employee rides ahead of a 2026 public launch. Tesla launched a Robotaxi-branded pilot in Austin in June 2025 and a service in the San Francisco Bay Area later in the year, though most cars still carry a human supervisor as of late 2025. Amazon-owned Zoox and Baidu's Apollo Go (the dominant operator in China) round out the global fleet.^[7]

A notable architectural shift in 2025 was the convergence of approaches: Waymo, Tesla, and UK-based Wayve all now use end-to-end foundation models trained on driving data, narrowing the long-running gap between Waymo's modular sensor-fusion stack and Tesla's vision-only approach.^[7]

AI is also used to optimize freight, last-mile delivery, ride matching (Uber, Lyft, DoorDash), and air traffic flow. UPS's ORION routing system, which has used machine learning since 2013, is credited with cutting roughly 100 million miles per year from delivery routes.

Scroll sideways for more →

How is AI used in education?

The most widely deployed educational AI product as of 2026 is Khanmigo, an AI tutor and teaching assistant built by Khan Academy on top of GPT-4. Khanmigo guides students through problems with Socratic prompts rather than handing them answers. In 2024 a Microsoft partnership made Khanmigo for Teachers free to all K-12 educators in the United States, and the tool has since launched in India, Spanish-speaking Latin America, and parts of Europe and Asia.^[8]

Duolingo Max, Speak (an English conversation tutor), and Quizlet's Q-Chat fill out the consumer-facing tier. Universities and school districts also use AI for plagiarism and AI-content detection, automated short-answer grading, and learning analytics to flag students at risk of falling behind. (See AI in education for the full overview.)

Scroll sideways for more →

How is AI used in software development?

AI coding tools have become one of the fastest-adopted enterprise technologies in recent memory. GitHub Copilot reached 20 million cumulative users in July 2025 and is in use at roughly 90% of Fortune 100 companies. Cursor (built on top of Anysphere's fork of VS Code) captured about 18% of the AI coding market within 18 months of launch, surpassed 1 million daily active users in 2025, and had developer users at more than half of the Fortune 500 by its Series C funding round in June 2025. The broader AI coding assistant market grew from $4.91 billion in 2024 to $7.37 billion in 2025.^[9]

GitHub reports that AI tools now generate roughly 46% of the code committed by developers using Copilot, and a 2025 Pragmatic Engineer survey found 85% of respondents use at least one AI coding tool. Sentiment is mixed; positive developer sentiment toward AI tools fell from 70% in 2024 to 60% in 2025 as users came up against the hallucinations and edge cases that the tools still struggle with.^[9]

Scroll sideways for more →

How is AI used in customer service?

Contact centers were among the first business functions hit hard by large language models. The global chatbot market sat at roughly $9.56 billion in 2025 and is expected to reach about $25.88 billion by 2030, while the broader Contact Center as a Service market was valued at $7.08 billion in 2025 and is projected to hit $30.15 billion by 2034.^[10]

Around 80% of companies are using or planning to use AI-powered chatbots for customer service, although only about 25% of call centers report that they have actually integrated AI into daily workflows. IBM has reported that AI can reduce customer service operational costs by 30 to 50%, and Gartner has forecast that by 2026 conversational AI in contact centers will reduce agent labor costs by $80 billion globally. The current wave is dominated by agentic systems (Sierra, Decagon, Ada, Salesforce Agentforce) that can take actions like processing refunds or rescheduling appointments rather than just answering questions.^[10]

How is AI used in manufacturing and robotics?

Manufacturing is the heaviest user of AI in heavy industry. By 2025 roughly 90% of manufacturers reported using AI in some part of their operations, and the global AI-in-manufacturing market is projected to grow from $5.94 billion in 2024 to $60.7 billion by 2034.^[11]

The two biggest application clusters are predictive maintenance and machine vision. Predictive maintenance models ingest vibration, thermal, current, and acoustic sensor streams to flag bearing wear, motor degradation, or hydraulic leaks before they cause downtime. Reported downtime reductions of 50 to 70% are common in published case studies. On the robotics side, "physical AI" platforms such as NVIDIA Isaac, Covariant RFM, and Skild AI give industrial robots better grasping, path planning, and bin-picking in unstructured environments. Digital twins (running in NVIDIA Omniverse and similar tools) let manufacturers train robots in simulation before deploying them on the line.^[11][12] (See AI in manufacturing for more.)

Scroll sideways for more →

How is AI used in defense and security?

The U.S. Department of Defense has formalized Project Maven (announced in 2017) into a long-term Pentagon-wide program. In May 2025 the Maven Smart System contract ceiling was raised to $1.3 billion through 2029, up from $480 million. Maven uses computer vision to detect and identify objects in drone footage and satellite imagery and feeds those targets into analyst workflows. The Pentagon has credited Maven with supporting 2024 U.S. airstrikes in Iraq, Syria, and Yemen, as well as the location of hostile maritime assets in the Red Sea.^[13]

In December 2024 Palantir and Anduril announced a consortium that links Anduril's Lattice Mesh (frontline sensors and effectors) with Palantir's Maven Smart System and AI Platform, intended to move sensor data into AI-supported targeting workflows. NATO's Communications and Information Agency acquired the Maven Smart System NATO (MSS NATO) in March 2025 for use within Allied Command Operations.^[13]

AI is also used heavily in cybersecurity (anomaly detection, threat hunting, phishing detection) and in border and surveillance systems. The use of AI in lethal autonomous weapons, facial recognition for law enforcement, and mass surveillance is also one of the most active areas of policy debate in AI ethics.

How is AI used in scientific research?

AI for science is perhaps the area where the technology has produced the most unambiguously useful results so far. Three of the headline projects:

• AlphaFold 2 and 3 (Google DeepMind / Isomorphic Labs) cover protein structure and protein-ligand interactions; the AlphaFold Protein Structure Database contains over 214 million predicted structures and the underlying work won the 2024 Nobel Prize in Chemistry.^[4][5]
• GNoME (Graph Networks for Materials Exploration), also from DeepMind, used graph neural networks to predict 2.2 million stable inorganic crystals, of which 380,000 are predicted to be thermodynamically stable. Berkeley Lab's A-Lab robotic synthesis platform successfully made 41 of 58 predicted compounds during a 17-day autonomous run, a 71% hit rate.^[14]
• GraphCast (DeepMind) produces global 10-day medium-range weather forecasts in under a minute on a single TPU, beating the European Centre for Medium-Range Weather Forecasts' HRES system on most measures in 2023 benchmarks.

AI is also widely used in genomics (DeepVariant, Enformer), astronomy (galaxy classification, exoplanet detection in Kepler/TESS data), particle physics (jet tagging at the LHC), and climate modeling.

How is AI used in agriculture?

Precision agriculture pairs AI with sensors, satellite imagery, and autonomous vehicles. John Deere unveiled a new fleet of fully autonomous tractors in February 2025 with 16 cameras in pods around the machine for 360-degree vision and sub-inch GPS. Modern Deere tractors process more than 50 million data points per season, combining soil sensors, satellite imagery, and weather feeds to time irrigation, fertilization, and crop protection. Deere reports that its See & Spray system can reduce chemical herbicide use by up to 70% by spraying weeds individually instead of broadcast spraying entire fields, and that AI plus advanced analytics can lift yields by around 30% relative to non-AI baselines.^[15]

Other agricultural applications include disease detection from leaf imagery (Plantix, Taranis), livestock monitoring (Connecterra, Cainthus), greenhouse environment optimization, and weed-pulling robots like Carbon Robotics' LaserWeeder.

How is AI used in retail and e-commerce?

Recommendation systems at large e-commerce platforms have run on machine learning for two decades; Amazon has said that recommendations generate over 35% of its revenue. The newer generative AI layer is conversational shopping. Amazon's Rufus assistant served about 250 million shoppers in 2025 with a reported 60% conversion rate for users who engaged with it, and Amazon's "Help me decide" feature, built on Bedrock and SageMaker, uses an LLM plus an evaluator LLM to generate personalized product recommendations.^[16]

Walmart's Sparky, Shopify's Sidekick, Mercado Libre's Asistente, and Klarna's AI assistant cover similar ground for other retailers. Behind the scenes, AI handles dynamic pricing, demand forecasting, fraud detection, and computer vision for cashier-free stores (Amazon Go used Just Walk Out technology before being scaled back to Dash Carts in 2024).

How is AI used in the creative arts and media?

Generative AI in the creative industries grew explosively after the 2022 release of Stable Diffusion, DALL-E 2, Midjourney, and ChatGPT. The generative art and design market is projected to grow from about $298 million in 2023 to roughly $8.6 billion by 2033 (a ~40% CAGR), and the generative AI music market from $570 million in 2024 to about $2.79 billion by 2030 (~30% CAGR). Surveys suggest around 60% of working musicians already use AI in some part of their workflow, and 82% of listeners in one study could not tell AI-generated compositions from human ones.^[17]

Text-to-video systems (OpenAI Sora, Google Veo, Runway Gen-3, Kling) have moved from research demos to commercial pilots in advertising, film pre-visualization, and short-form social content. OpenAI released Sora to ChatGPT Plus and Pro subscribers in December 2024 and followed with Sora 2 in September 2025. AI is also used in dubbing (ElevenLabs, Papercup), de-aging, deepfake detection, and post-production color grading.

Applications by modality

The rest of the page lists representative AI applications grouped by input and output modality. Many tools in the table at the end of the article combine several modalities.

Content Generation (Text-to-Text)

• ChatGPT

• Jasper

• DeepMind Sparrow

• Writesonic

• Character.AI - Create characters and talk to them

Chatbots

• ChatGPT by OpenAI

• Chatsonic by Writesonic

• Claude by Anthropic

• Bard by Google

• Jasper Chat by Jasper

• ERNIE Bot by Baidu

Writing Assistants

• Jenni AI

• Pragma - sales messages

• Lex.page

• Craftly.AI - copywriting

• Copy.ai - write copy

• Rytr

• Microsoft 365 Copilot

Translation

• DeepL Translator

Image Generation (Text-to-Image)

• Midjourney

• Stable Diffusion

• DALL-E

• Lexica - Stable Diffusion search engine

• Adobe Firefly

• Microsoft Designer - Design powered by DALL-E

Video Generation (Text-to-Video)

• Synthesia - videos with AI avatars

• Lumen5 - turn blog posts into marketing videos

Audio Generation (Text-to-Audio)

• Murf

• Synthesys

• play.ht

• Prime Voice AI by Eleven Labs

Speech-to-Speech

• Resemble AI

Speech-to-Text

• Amazon Transcribe

• OpenAI Whisper

• AssemblyAI

• Descript

Image-to-Text

Amazon Textract

Video-to-Text

• Poised - communication coach

Image-to-Image

• Bria - Stockphoto manipulation

Image-to-Video

• D-ID - create speaking portrait videos

Video-to-Video

• Runway Gen-1

Search Engines

See also: Search Engines

• You.com

• Perplexity AI

• phind - developer focused

Music Creation

• AIVA - music composition

• Amper AI

• MusicML - text to music, Google

• Noise2Music - using diffusion models

Software Development

• GitHub Copilot X

• GitHub Copilot

• Amazon CodeWhisperer - free

• Tabnine

• Replit - online, multiplayer IDE

• CodeGen - Salesforce

AI Content Detectors

AI Content Detectors

Education

Language learning

• Speak

Games

• AI Dungeon

Other Text-based Apps

• Calligrapher.ai - AI handwriting generation Discussion: https://news.ycombinator.com/item?id=34530011

Mem.ai - automated workspace organization

• 24.5 million series A funding led by OpenAI

timelyAI - scheduling using Whatsapp Mutiny - website conversion, A/B testing, B2B DeepBeat - generate rap lyrics punchlines.ai - generates jokes Shortwave - email organization, summarization, tools Steeped AI - survey analysis, quant research

Other Image-based Apps

• Lensa AI - image (profile pictures) editing

• scenario.gg - game assets

• Segment Anything - crop out any object in an image with 1 click

• Leonardo.Ai - game assets

Other Video-based App

• Repurpose.io - automatically repurpose existing video content to other social media platforms

• Eightify - summarizes YouTube videos

• Wonder Studio - add CG characters to live action scenes

Design

• Spline AI - 3D Design

Biology and Medicine

• AlphaFold - predicts protein structure

• Atomic AI - RNA drug discovery

Model Deployment

• NVIDIA Triton Inference Server

• NVIDIA Picasso - build and deploy generative AI-powered image, video, and 3D apps

• replicate - use open-source models with a Cloud API

• Azure OpenAI Service

• Banana.dev - deploy and use open source ML models

• Amazon Bedrock

Model Training

• Amazon Bedrock

Finance

• Broker Compare

Table

See also: AI apps table

Scroll sideways for more →

References

1. The Imaging Wire, "FDA AI Approvals Surge Past 1k for Radiology" (Dec 2025). https://theimagingwire.com/2025/12/10/ai-enabled-medical-devices-granted-fda-marketing-authorization/ ↩
2. U.S. Food and Drug Administration, "Artificial Intelligence and Machine Learning (AI/ML)-Enabled Medical Devices." https://www.fda.gov/medical-devices/software-medical-device-samd/artificial-intelligence-enabled-medical-devices ↩
3. Philips, "Philips Advances MRI Speed and Precision with FDA 510(k) Clearance of SmartSpeed Precise Dual AI Software" (2025). https://www.philips.com/a-w/about/news/archive/standard/news/articles/2025/philips-advances-mri-speed-and-precision-with-fda-510k-clearance-of-smartspeed-precise-dual-ai-software.html ↩
4. Varadi, M. et al., "AlphaFold Protein Structure Database in 2024," Nucleic Acids Research. https://academic.oup.com/nar/article/52/D1/D368/7337620 ↩
5. Google DeepMind, "AlphaFold 3 predicts the structure and interactions of all of life's molecules" (2024). https://blog.google/innovation-and-ai/products/google-deepmind-isomorphic-alphafold-3-ai-model/ ↩
6. RGP, "AI in Financial Services 2025." https://rgp.com/research/ai-in-financial-services-2025/ ↩
7. CNBC, "Waymo, Zoox and Tesla drive 2025 robotaxi boom" (Dec 2025). https://www.cnbc.com/2025/12/16/waymo-amazon-zoox-tesla-robotaxi-expansion.html ↩
8. Khan Academy, "Khanmigo: Khan Academy's AI-powered teaching assistant & tutor." https://www.khanmigo.ai/ ↩
9. Second Talent, "GitHub Copilot Statistics & Adoption Trends 2026." https://www.secondtalent.com/resources/github-copilot-statistics/ ↩
10. Nextiva, "50+ Conversational AI Statistics for 2026." https://www.nextiva.com/blog/conversational-ai-statistics.html ↩
11. World Economic Forum, "What is physical AI, and how is it changing manufacturing?" (Sept 2025). https://www.weforum.org/stories/2025/09/what-is-physical-ai-changing-manufacturing/ ↩
12. A3 Automate.org, "Industrial AI in Action: Predictive Maintenance and Operational Efficiency at Scale." https://www.automate.org/blogs/industrial-ai-in-action-predictive-maintenance-and-operational-efficiency-at-scale ↩
13. Wikipedia, "Project Maven." https://en.wikipedia.org/wiki/Project_Maven ↩
14. Google DeepMind, "Millions of new materials discovered with deep learning" (2023-2024). https://deepmind.google/blog/millions-of-new-materials-discovered-with-deep-learning/ ↩
15. Farmonaut, "Ag Power John Deere: AI Products, 2025 Farming Trends." https://farmonaut.com/precision-farming/ag-power-john-deere-ai-products-2025-farming-trends ↩
16. TechCrunch, "Amazon's new AI shopping tool tells you why you should buy a recommended product" (Oct 2025). https://techcrunch.com/2025/10/23/amazons-new-ai-shopping-tool-tells-you-why-you-should-buy-a-recommended-product/ ↩
17. ArtSmart.ai, "AI in Music Industry Statistics 2025: Market Growth & Trends." https://artsmart.ai/blog/ai-in-music-industry-statistics/ ↩
18. Wikipedia, "Applications of artificial intelligence." https://en.wikipedia.org/wiki/Applications_of_artificial_intelligence
19. Stanford Institute for Human-Centered AI, "The 2025 AI Index Report" (Economy chapter). https://hai.stanford.edu/ai-index/2025-ai-index-report/economy ↩
20. McKinsey & Company, "The state of AI in 2025: Agents, innovation, and transformation." https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai ↩
21. The Imaging Wire, "Numbers from the FDA Show Radiology Is Maintaining Its Lead" (Mar 2026). https://theimagingwire.com/2026/03/11/numbers-from-the-fda-show-radiology-is-maintaining-its-lead/ ↩
22. Aidoc, "Aidoc Secures Landmark FDA Clearance for First Foundation Model-Powered Clinical AI Solution of Its Kind" (Feb 2025). https://www.aidoc.com/about/news/aidoc-secures-landmark-fda-clearance/ ↩
23. AI Magazine, "How JPMorgan is Embedding AI into Employee Performance" (2025). https://aimagazine.com/news/how-jpmorgan-is-embedding-ai-into-employee-performance ↩
24. CNBC, "Waymo crosses 450,000 weekly paid rides as Alphabet robotaxi unit widens lead on Tesla" (Dec 2025). https://www.cnbc.com/2025/12/08/waymo-paid-rides-robotaxi-tesla.html ↩