AI use cases
2983 ways people actually use AI, compiled from the videos in this
library. Each names the job and why it is worth doing, and links to the
moment in the source video.
-
Review and validate outputs through multiple AI systems
— Evaluate accuracy, consistency, novelty, and grounding before synthetic data enters future training pipelines.
-
Retrieve fresh information through retrieval augmented generation
— Allow models to consult external sources instead of relying solely on recursively learned patterns.
-
Generate synthetic text, images, code, videos, and social media posts
— Create content that may later enter AI training datasets, while introducing risks of recursive training and model collapse.
-
Handle a conversational contact workflow and route submitted information to an inbox
— Capture customer requests such as pizza preferences, delivery choice, and dietary requirements
-
-
Generate complete business websites from natural-language prompts
— Create an initial website structure and design without traditional development work
-
Suggest search terms and agentic use cases
— Identify potential article topics and ways to apply AI agents to daily activities, work, hobbies, or business.
-
Generate website features and content
— Add article tagging, author information, internal links, and SEO-focused articles.
-
Generate a website from prompts
— Create and launch a news website without manually coding its design and functionality.
-
Operate across connected messaging, email, calendar, and other applications
— Provide a personal AI assistant capable of taking actions on the internet.
-
Create a website from a single prompt
— Launch a business website quickly.
-
Generate possible customer search terms for forestry mulching and land clearing
— Expand the list of consumer-oriented queries before running them through a keyword tool.
-
Automated spam defense
— An AI agent replies to unsolicited outreach, asks questions, schedules calls, requests follow-ups and contracts, and wastes the sender's time in return.
-
Mass personalized outreach
— AI sales development representatives scrape contact information and send emails, texts, and calls that appear to come from real people.
-
Generating a purpose, tagline, audience definition, and channel positioning
— Vanessa used ChatGPT AI to try to define her mission and reverse-engineer her identity, platform behavior, habits, and audience.
-
Help determine metrics and how to find them
— Support the creation and use of business scorecards.
-
Review calendars, email, and business information
— Identify top priorities, summarize the schedule, find time-consuming activities, suggest where to refocus energy, identify emails requiring responses, and draft replies.
-
Build an interactive newsletter-generation app
— Capture ideas during the week, structure the newsletter, and reduce writing time from four hours to less than an hour.
-
Create a promotions calendar and Notion database
— Plan email promotions around US holidays and pop-culture holidays related to tea, bubble tea, and milk tea.
-
Transcribe meetings and extract summaries and action items
— Avoid taking meeting minutes and allow full participation in meetings.
-
-
Read websites and deliver useful answers without requiring a site visit.
— Compress the human research and purchasing process into an agent-generated answer.
-
Retrieve data, compare products, call tools, and take actions on behalf of users.
— Use the internet as infrastructure to complete tasks rather than merely direct users to pages.
-
Self-improvement
— Use feedback and evaluation tests to identify errors, improve accuracy, and adjust the agent's personality and verbosity.
-
Proactive business monitoring
— Review social media, parse keywords, track product metrics, and send scheduled reports without waiting for a prompt.
-
Content creation
— Draft marketing materials, engineering blog posts, product updates, and other business communications.
-
Company knowledge and data analysis
— Answer questions about customers, internal systems, company practices, metrics, and organizational expertise.
-
Coding and software development
— Enhance difficult engineering work, build prototypes and applications, and automate knowledge-worker workflows.
-
Research and data review
— Review substantially more data and help researchers investigate problems that are difficult to fund or solve manually.
-
Model distillation and reinforcement learning
— Use outputs from openweight models to teach or improve another model, subject to applicable terms of service.
-
Sub-agent orchestration
— Use low-cost models for deterministic subtasks while an orchestrator model handles broader, less predictable work.
-
PII redaction
— Add a privacy and safety layer to inference deployed across an enterprise.
-
Prompt-injection detection
— Flag prompts that appear to contain prompt-injection attempts.
-
Continuous benchmarking
— Measure how different models perform across inference providers and track changing results.
-
Model routing and provider selection
— Improve quality, speed, price and uptime by directing traffic toward better-performing inference providers.
-
Agent evaluation and improvement
— Evals connect agent behavior to business outcomes and provide feedback for training and iteration.
-
Vehicle inspection and repair support
— The El Mike agent acts as a mechanic's sidekick by providing inspection guidance and tips.
-
Executive decision-making
— An AI CEO made forecasts, managed a city's operations, assigned daily plans, and monitored performance.
-
Financial services
— Agents support car-loan approval, underwriting, pricing, servicing, and personalized loan offers.
-
Customer relationship management and sales
— Agents remember customer history, develop strategies, pursue long-term goals, and sell cars and related products.
-
Analyze Ethereum L1 transaction volume.
— Assess the rise in Ethereum L1 activity and the contribution from TokenWorks.
-
Analyze gold and silver performance during the last week.
— Provide a second opinion on the metals' recent price movements and catalysts.
-
Compare US stock-market performance under different presidents.
— Assess whether Trump's stock-market performance was unusually strong.
-
Generate an HTML report about SpaceX.
— Turn a market discussion into a shareable report containing its stated revenue, compute targets, capital expenditures and risks.
-
-
Execute routine steps and scripts
— The agent handles routine work while deterministic scripts perform fragile logic that should not be improvised.
-
Select whether to run a skill
— The agent uses each skill's name and description at startup to decide which skill to trigger.
-
Perform a specific job using procedural instructions
— An agent skill supplies the particular process and expertise that the model does not already know.
-
Vulnerability remediation workflow automation
— The Lightwell Engine uses AI assistance within an approximately 99% automated process for creating library artifacts.
-
Threat hunting
— AI can execute a human-designed hunting plan, investigate threats more deeply, and reduce the manual burden of triage and enrichment.
-
Finding, explaining, and chaining software vulnerabilities
— AI models can assist offensive and defensive cybersecurity work, including vulnerability discovery and identifying combined effects from multiple vulnerabilities.
-
Training models for specific capabilities
— Companies select capabilities, collect relevant data and use human contractors to annotate and train models for those capabilities.
-
Autonomous driving
— AI models identify vehicles, pedestrians, traffic lights and lane markings so driving software can act autonomously.
-
Supporting medical diagnosis
— AI used by radiologists as an input to human judgment was described as improving the accuracy and early detection of certain cancers.
-
Automating customer-service conversations
— Klarna's CEO said AI handled 70% of the company's customer-service conversations.
-
Mechanistic interpretability
— Researchers use interpretability methods to understand how trained neural networks process information and make decisions.
-
Robotics and physical labor
— Superintelligent AI could operate robots that perform physical tasks, build factories and expand the use of AI throughout the economy.
-
Automating AI research
— AI systems are intended to generate research ideas, analyze experiments, communicate results and eventually perform the entire research process autonomously.
-
Autonomous coding
— AI companies are training systems to write and edit code so they can accelerate their own work and AI progress.
-