nicornleCategory
30 questions
- ·By end of 2025, what percentage of Fortune 500 companies had deployed AI in at least one production business process?Adoption→
- ·Meta's Chief AI Scientist Yann LeCun (Turing Award winner) has a controversial position on LLMs. What is it?People→
- ·Anthropic's Dario Amodei and OpenAI's Sam Altman were once colleagues. Where did Dario work before founding Anthropic?People→
- ·At Nvidia's peak in 2025, what was Jensen Huang's estimated net worth?People→
- ·In February 2026, which AI company accused DeepSeek of scraping its outputs to train their model?Industry→
- ·In 2025, a well-known AI coding agent caused a real production incident. What did it do?Security→
- ·Prompt injection is OWASP's #1 LLM security risk. What is the typical attack success rate against production AI systems?Security→
- ·By 2026, what do most senior AI engineers recommend for new production AI applications?Tools→
- ·Anthropic's Claude Code launched in early 2025. What annualised revenue run rate did Anthropic hit by mid-2026?Revenue→
- ·Anthropic released MCP (Model Context Protocol) in late 2024. How many MCP servers exist by mid-2026?Tools→
- ·In a January 2026 developer survey (JetBrains AI Pulse), which AI coding tool had the highest workplace usage?Tools→
- ·A model claims a 1M token context window. At what actual token length does accuracy typically start degrading?Architecture→
- ·What is the #1 reason production AI agents fail on long multi-step tasks in 2026?Architecture→
- ·For most enterprise AI deployments in 2025, which approach delivered better production accuracy?Architecture→
- ·DeepSeek achieved frontier performance at 1/17th the training cost. What architecture made it possible?Architecture→
- ·Running 1 million GPT-4 equivalent queries cost ~$30,000 in 2023. What does the same workload cost in 2026?Pricing→
- ·Nvidia's H100 GPU costs ~$30K. How many tokens per second can a single H100 generate at inference?Infrastructure→
- ·OpenAI, SoftBank and Oracle announced Stargate — a US AI infrastructure project. What is the total committed investment?Infrastructure→
- ·DeepSeek R1 vs OpenAI o1: how much cheaper is DeepSeek R1 per million tokens?Pricing→
- ·In enterprise AI deployments in 2025, what percentage of AI responses contained factual errors or hallucinations?Risk→
- ·On SWE-Bench Verified (real GitHub issues), which AI coding model scores highest in 2026?Models→
- ·Since 2023, how much has the price per 1M output tokens fallen across leading LLMs?Pricing→
- ·OpenAI spent ~$100M training GPT-4. DeepSeek matched its performance. What did DeepSeek's training cost?Efficiency→
- ·In 2023, GPT-4 launched with a context window of ~8K tokens. What is the standard frontier context window in 2026?Models→
- ·In fiscal year 2025, how much revenue did Nvidia make from AI data centres alone?Revenue→
- ·Cursor, the AI coding editor, went from $0 to $1B in annualised revenue. How long did it take?Growth→
- ·In 2026, how much do Amazon, Microsoft, Google + Meta plan to spend — combined — on AI data centres?Infrastructure→
- ·Stripe's valuation hit an all-time high in February 2026. What was it?Valuation→
- ·In May 2026, Anthropic overtook OpenAI as the world's most valuable AI startup. What valuation did it reach?Valuation→
- ·OpenAI just closed the largest private funding round in tech history. How much did they raise in a single round?Funding→