Explore 94,000 AI Use Cases

Cutting through the AI noise

Welcome to the Women Who AI Newsletter, your weekly update on what actually matters in AI when you’re focused on building and scaling startups.

Was this forwarded to you? We send out newsletters every Monday morning. Click here to subscribe and join our community of founders shaping the future of AI.

Deep Dive: 94k LLM Use Cases

The REALM Dataset was released last week. This impressive work aggregates 94,000 examples of how different large language models (LLMs) are being used across various professions and how that has changed over time.

The data was collected from Reddit and news articles between June 2020 (when GPT-3 was released) and December 2024. Read more about their methodology here or on arXiv.

Key Insights

• Sources: 78,749 news articles, 14,510 Reddit posts

• Time period: 2020 to 2024

• Growth: 4,370.7% increase from 2020 to 2024

• Top use cases: Content Synthesis, Content Creation, and Process Automation

• Industries: Healthcare, Legal, and Education sectors show rapidly increasing adoption in recent periods

Explore Use Cases

Our team has created an interactive dashboard to help you explore this data, including ways to analyze use cases across time and different professions:

Insights For Entrepreneurs Building With LLMs

Our analysis of the REALM dataset reveals several strategic opportunities for founders:

Target High-Growth Areas: Content creation, summarization, decision support, digital assistance, and process automation show the strongest adoption trends—making them prime opportunities for new ventures.

Industry-Specific Markets: Art, media, business, computer science, and management professionals demonstrate the highest LLM engagement, suggesting they’re ready for specialized tools.

Discover Untapped Potential: Sectors like robotics automation, recommendation systems, and vehicular automation have minimal LLM integration, potentially indicating blue ocean opportunities.

Education Sector Applications: The data highlights concrete demands in educational settings for lesson planning, quiz creation, rubric design, and administrative support.

Insights For Professionals Looking to Apply LLMs to Their Work

If you're looking to enhance your productivity with LLMs, consider these evidence-based approaches:

Content Professionals (Arts & Media): Content creation and summarization tools show the highest established value. These professionals are consistently the most engaged LLM users across both news sources and Reddit.

Business & Management: Implement LLMs for decision-making tasks such as business strategy development, marketing campaign optimization, and financial investment guidance—all areas showing substantial growth.

Computer Science & Mathematics: Beyond coding assistance, utilize LLMs for discovery-oriented tasks to advance research and innovation.

Process Automation: Identify your most time-consuming routine tasks as prime candidates for LLM automation—the research shows significant interest in applications that streamline repetitive work.

Decision Support: LLMs excel at recommendation-based tasks, helping synthesize options and suggest optimal approaches across business, marketing, and financial domains.

Last Week’s Generative AI Highlights

Last week, we covered Gemini 2.0's image generation and editing capabilities and how the technology might work under the hood.

Since then, releases have dropped rapidly. In response to Gemini's release, OpenAI launched its own improved image generation and editing features in ChatGPT, sparking significant attention (and some controversy).

Capabilities of OpenAI’s GPT 4o image editor:

• Complex Multi-Object Prompting: Handle up to 20 separate objects in a single image generation prompt

• Background Transparency: Easily make image backgrounds transparent

• Typography Control: Specify font styles, sizes, and color schemes for text elements in generated images

• Advanced Photo Editing: Remove reflections, enhance color grading, or add weather effects (like rain)

• Style Reference: Upload images to use as style guides for your generations

These are available through natural language prompting directly in the existing ChatGPT 4o interface. Read more in the OpenAI official announcement.

At the same time, Halfmoon (officially known as Reve Image) quietly dropped a new model, which tops the Artificial Analysis Image Arena Leaderboard. It can be tested for free today. Just input what you want to see—no prompt engineering needed.

Try it here: preview.reve.art

Google has also released Gemini 2.5, which brings significant updates to its flagship AI model. It has a massive one-million token context window and multimodal reasoning that works across documents and visuals.

Gemini 2.5 Pro is currently available through Google AI Studio and will be available on Vertex AI soon.

Hackathons

Ready to build that product you've been dreaming about? Check out these upcoming hackathons!

If you'd like to find a Women Who AI team for any event, reply to this email, and we'll connect everyone interested.

LLM x Law Hackathon @Stanford #5 | San Francisco | Sun April 6 | Event Link

AI Agents Hackathon by Microsoft | Virtual | April 8-30 | $50K in prizes | Event Link

Peregrine: Emergency Response Tech Hackathon | Boston | Sat April 12 | Event Link

Robot Arm Hackathon | NYC | Sat & Sun April 19-20 | Event Link

Jobs We Love: Internships with Wing VC Startups

Wing VC, a leading venture capital firm with a partner specializing in healthcare and biotech investments, has portfolio companies spanning multiple sectors. Several Wing VC portfolio companies are currently seeking summer interns across Growth, Operations, Product, and Engineering.

To Apply:

1. Submit your LinkedIn profile through this form

2. Submissions will be shared directly with Wing's head of portfolio talent

3. Portfolio companies will contact qualified candidates directly if there's a potential match