| 90% | “90%” | Kant: Model capabilities determine 80-90% of AI product value | Eiso Kant | Dec 20, 2023 | — |
| 100% | “hundred percent” | Sapoznik: Call center agents face an annual turnover rate near 100% | Gustavo Sapoznik | Dec 13, 2023 | Contradicted |
| 4M | “four million” | Sapoznik: The US employs roughly 3 to 4 million call center agents | Gustavo Sapoznik | Dec 13, 2023 | Supported |
| 80% | “80%” | Sapoznik: Agents click ASAPP's AI suggestions 80% of the time | Gustavo Sapoznik | Dec 13, 2023 | — |
| 1% | “one percent” | Sapoznik: Under 1% of traditional call center interactions undergo manual QA | Gustavo Sapoznik | Dec 13, 2023 | — |
| 10% | “10%” | Katzenberg: Manual vendor spend audits consistently yielded about 10% savings | Howard Katzenberg | Nov 29, 2023 | — |
| $15M | “fifteen million dollars” | Better identified $1.5M in savings on $15M of annual vendor spend | Howard Katzenberg | Nov 29, 2023 | — |
| 1.5M | “1.5 million” | Better identified $1.5M in savings on $15M of annual vendor spend | Howard Katzenberg | Nov 29, 2023 | — |
| 100% | “hundred percent” | Glean AI shifted 100% of its machine learning stack to LLMs | Howard Katzenberg | Nov 29, 2023 | — |
| 3% | “three percent” | League Apps estimates Glean AI saves 2-3% of non-payroll spend | Howard Katzenberg | Nov 29, 2023 | — |
| 3× | “three X” | Katzenberg: Glean AI business grew over 3x in 2023 | Howard Katzenberg | Nov 29, 2023 | — |
| 30% | “30%” | Bapu: 30% of CMIOs are evaluating and purchasing generative AI | Akilesh Bapu | Nov 17, 2023 | Partly supported |
| 20× | “20 x” | Bapu: HLTH conference attendance expanded 20x from 2019 to 2023 | Akilesh Bapu | Nov 17, 2023 | Partly supported |
| 1.5M | “a half million” | DeepScribe has gathered over 2.5 million labeled medical conversations | Akilesh Bapu | Nov 17, 2023 | — |
| 50% | “50%” | Bapu: Only 50% of oncologist documentation comes from direct conversations | Akilesh Bapu | Nov 17, 2023 | — |
| 6M | “six million” | DeepScribe grew from zero to $6M ARR purely through inbound marketing | Akilesh Bapu | Nov 17, 2023 | — |
| 100% | “hundred percent” | DeepScribe achieves nearly 100% patient consent rate | Akilesh Bapu | Nov 17, 2023 | — |
| 1B | “a billion” | Zhou: Fine-tuning transformed GPT-3 into ChatGPT | Sharon Zhou | Nov 8, 2023 | Supported |
| 100B | “a hundred billion” | Zhou: Lamini can train models up to 100 billion parameters | Sharon Zhou | Nov 8, 2023 | — |
| 1K | “a thousand” | Lamini switches across 1,000 fine-tuned models in three milliseconds | Sharon Zhou | Nov 8, 2023 | Open |
| 7% | “seven percent” | Lodge: DeepMind shaved 7% off C++ sort algorithm using reinforcement learning | Mathew Lodge | Oct 25, 2023 | Partly supported |
| 100B | “hundred billion” | Diffblue generates unit tests in 1.5 seconds versus 40 seconds for LLMs | Mathew Lodge | Oct 25, 2023 | — |
| 1B | “one billion” | Moonhub indexes data on one billion people across the public web | Nancy Xu | Oct 18, 2023 | — |
| 80% | “80%” | Moonhub clients report 80% of candidates are unique to the platform | Nancy Xu | Oct 18, 2023 | — |
| 100M | “a hundred million” | Over 100 US recruiting agencies generate more than $100 million annually | Nancy Xu | Oct 18, 2023 | Supported |
| 1K | “a thousand” | The prompt engineer title jumped to roughly 1,000 SF workers in 2023 | Nancy Xu | Oct 18, 2023 | — |
| 10× | “10 times” | Polu: Building a strong AI research team is 10x easier in Paris than SF | Stanislas Polu | Oct 11, 2023 | — |
| 95% | “95%” | Kanjun Qiu: 95% of Imbue's work is engineering, not research | Kanjun Qiu | Oct 4, 2023 | — |
| 100B | “hundred billion” | Qiu: Imbue pre-trains 100B+ parameter models from scratch | Kanjun Qiu | Oct 4, 2023 | — |
| 95% | “95%” | Goshen: 95% of enterprise AI use cases are task-specific | Ori Goshen | Sep 20, 2023 | — |
| 10M | “ten million” | Goshen: AI21 Labs' Wordtune has over 10 million users | Ori Goshen | Sep 20, 2023 | Supported |
| 5% | “five percent” | Taylor: Community building provides more startup growth leverage than marginal technical improvements | Carly Taylor | Sep 14, 2023 | — |
| 10× | “10 X” | Taylor: Community building provides more startup growth leverage than marginal technical improvements | Carly Taylor | Sep 14, 2023 | — |
| $45M | “forty-five million dollars” | Deepset announced a Balderton-led Series B, reaching $45 million total funding | Matt Turck | Sep 6, 2023 | Partly supported |
| 20% | “20%” | Shah: The US healthcare system is 20% to 30% understaffed in nursing | Munjal Shah | Aug 23, 2023 | Partly supported |
| 30% | “30%” | Shah: The US healthcare system is 20% to 30% understaffed in nursing | Munjal Shah | Aug 23, 2023 | Partly supported |
| 1.5T | “1.5 trillion” | Shah: Hippocratic AI acquired 1.5 trillion healthcare tokens for training | Munjal Shah | Aug 23, 2023 | — |
| 100% | “hundred percent” | Shah: Hippocratic AI operates fully in person five days a week | Munjal Shah | Aug 23, 2023 | — |
| 15% | “15%” | Biewald: No single industry accounts for over 15% of W&B revenue | Lukas Biewald | Aug 9, 2023 | — |
| 99% | “99%” | Biewald estimates 99% of Global 2000 use ML for core operations | Lukas Biewald | Aug 9, 2023 | — |
| 90M | “ninety million” | Matt Turck: Ada raised over $190 million in VC funding | Matt Turck | Jul 26, 2023 | Supported |
| 30M | “thirty million” | Matt Turck: Ada raised over $190 million in VC funding | Matt Turck | Jul 26, 2023 | Supported |
| 40% | “40%” | Customer service agent attrition rates exceed 40% normally, 80% during pandemic | Mike Murchison | Jul 26, 2023 | Supported |
| 80% | “80%” | Customer service agent attrition rates exceed 40% normally, 80% during pandemic | Mike Murchison | Jul 26, 2023 | Supported |
| 30% | “30%” | Generative AI replies increase customer service resolution rates by 30 percentage points | Mike Murchison | Jul 26, 2023 | — |
| 1B | “a billion” | Synthesia reached a $1B valuation in Series C led by Accel | Matt Turck | Jul 19, 2023 | Supported |
| 35% | “35%” | Riparbelli: Synthesia has over 50,000 customers and 35% of Fortune 100 | Victor Riparbelli | Jul 19, 2023 | Supported |
| 95% | “95%” | Riparbelli: LLMs are not production-ready for 95% of intended tasks | Victor Riparbelli | Jul 19, 2023 | — |
| 8.5M | “8.5 million” | LlamaIndex announced an $8.5 million seed round led by Greylock | Matt Turck | Jul 12, 2023 | Supported |
| 100M | “a hundred million” | Turck: Companies with $100M+ ARR have been built solely on data connectors | Matt Turck | Jul 12, 2023 | Supported |
| 15% | “15%” | Jerry Liu: 'Drop everything' reactiveness accounts for 10-15% of LlamaIndex's work | Jerry Liu | Jul 12, 2023 | — |
| 200M | “two hundred million” | Dataiku reached over $200 million ARR and 1,200 employees | Florian Douetteau | Jul 5, 2023 | Partly supported |
| 90% | “90%” | Sivulka: Multi-step AI workflows suffer from compounding accuracy errors | George Sivulka | Jun 21, 2023 | — |
| 10% | “10%” | Sivulka: Multi-step AI workflows suffer from compounding accuracy errors | George Sivulka | Jun 21, 2023 | — |
| 90% | “90%” | Sivulka: LLMs excel at first-order tasks but struggle with nth-order | George Sivulka | Jun 21, 2023 | — |
| 72% | “72%” | Hebbia removed cosine similarity scores after finding they confused users | George Sivulka | Jun 21, 2023 | — |
| 8B | “a seven billion” | Ratner: $200 of ChatGPT calls can clone closed models into open ones | Alex Ratner | May 31, 2023 | Partly supported |
| 1M | “a million” | Liberty: Even naive RAG implementation significantly reduces AI hallucinations | Edo Liberty | May 31, 2023 | — |
| 2B | “two billion” | AssemblyAI has processed almost two billion audio files to date | Dylan Fox | May 1, 2023 | — |
| 1K | “a thousand” | Fox: AssemblyAI serves over 1,000 customers and tens of thousands of monthly developers | Dylan Fox | May 1, 2023 | — |
| 4M | “four million” | AssemblyAI is training its next speech model on ~4 million hours of audio | Dylan Fox | May 1, 2023 | — |
| 100M | “a hundred million” | AssemblyAI processes over 100 million audio files monthly via API | Dylan Fox | May 1, 2023 | — |
| 15% | “15%” | State-of-the-art speech recognition models still carry a 15% error rate | Dylan Fox | May 1, 2023 | Supported |
| 10× | “10 times” | AssemblyAI's JAX contribution sped up Whisper model training by 10x | Dylan Fox | May 1, 2023 | Contradicted |
| 1% | “one percent” | Falcon: Probably only 1% of emerging generative AI tools will endure | William Falcon | May 1, 2023 | — |
| 100× | “hundred times” | Retrieval-augmented generation is 100 times cheaper than fine-tuning, says Amr Awadallah | Amr Awadallah | May 1, 2023 | — |
| 1K | “a thousand” | Top developers, lawyers, and doctors will become 1,000x more productive using AI | Amr Awadallah | May 1, 2023 | — |
| 10× | “10 X” | Top developers, lawyers, and doctors will become 1,000x more productive using AI | Amr Awadallah | May 1, 2023 | — |
| 100× | “hundred X” | Top developers, lawyers, and doctors will become 1,000x more productive using AI | Amr Awadallah | May 1, 2023 | — |
| 1K | “a thousand” | Reducing database query latency to 100ms cuts infrastructure requirements 10x | Jorge Gomez Sancha | Mar 9, 2023 | — |
| 10× | “10 times” | Reducing database query latency to 100ms cuts infrastructure requirements 10x | Jorge Gomez Sancha | Mar 9, 2023 | — |
| 40% | “40%” | Mann: GitHub Copilot saves developers about 40 percent time on new code | Gideon Mann | Feb 2, 2023 | Partly supported |
| 5M | “five million” | BigID generated $5M in revenue in year one and $15M in year two | Dimitri Sirota | Jan 30, 2023 | — |
| 15M | “fifteen million” | BigID generated $5M in revenue in year one and $15M in year two | Dimitri Sirota | Jan 30, 2023 | — |
| 40% | “40%” | CISOs and Chief Data Officers each drive 40% of BigID's business | Dimitri Sirota | Jan 30, 2023 | — |
| 20% | “20%” | CISOs and Chief Data Officers each drive 40% of BigID's business | Dimitri Sirota | Jan 30, 2023 | — |
| 100M | “a hundred million” | BigID expects to cross $100 million in revenue during 2023 | Dimitri Sirota | Jan 30, 2023 | — |
| $1M | “a million dollars” | BigID generated nearly $1 million from its SmallID product since September 2022 | Dimitri Sirota | Jan 30, 2023 | — |
| 1K | “a thousand” | Kimball: Google ran over 1,000 MySQL shards for a decade before Spanner | Spencer Kimball | Nov 17, 2022 | Partly supported |
| 40M | “forty million” | Berkowitz: ADP pays about 40 million workers globally every month | Jack Berkowitz | Nov 17, 2022 | Partly supported |
| $3.2T | “3.2 trillion dollars” | Jack Berkowitz says ADP moved $3.2 trillion in money transactions last year | Jack Berkowitz | Nov 17, 2022 | Partly supported |
| 40M | “forty million” | ADP uses ML to reduce 21 million job titles into 7,000 categories | Jack Berkowitz | Nov 17, 2022 | — |
| 21M | “twenty one million” | ADP uses ML to reduce 21 million job titles into 7,000 categories | Jack Berkowitz | Nov 17, 2022 | — |
| 95% | “95%” | ADP's skills graph ontology is 95% automated | Jack Berkowitz | Nov 17, 2022 | — |
| 20% | “20%” | ADP pays roughly 20 percent of all workers in the United States | Jack Berkowitz | Nov 17, 2022 | Partly supported |
| 30% | “30%” | Berkowitz: ADP pays 20% to 30% of the French population | Jack Berkowitz | Nov 17, 2022 | — |
| 18% | “18%” | Berkowitz: ADP handles recruiting for 15% to 18% of US workers | Jack Berkowitz | Nov 17, 2022 | Contradicted |
| 4B | “four billion” | Alex Dean: Strava generates four billion events per day using Snowplow | Alex Dean | Oct 24, 2022 | — |
| 100M | “a hundred million” | Turck: Fivetran generates over $100M in annual revenue | Matt Turck | Oct 10, 2022 | Supported |
| 1K | “a thousand” | Stancil: Enterprise data market won't support 1,000 unicorns | Benn Stancil | Oct 10, 2022 | — |
| 70% | “70%” | Kim: Analysts and engineers make up 70% of Select Star's user base | Shinji Kim | Sep 12, 2022 | — |
| 26M | “a twenty-five million” | Mezhanskiy: Datafold's data-diff benchmarks over 1B rows in 5 minutes | Gleb Mezhanskiy | Sep 12, 2022 | Not publicly verifiable |
| 1B | “one billion” | Mezhanskiy: Datafold's data-diff benchmarks over 1B rows in 5 minutes | Gleb Mezhanskiy | Sep 12, 2022 | Not publicly verifiable |
| $1.5B | “1.5 billion dollars” | Moses cites 2022 revenues for BigQuery, Snowflake, and Databricks | Barr Moses | Apr 12, 2022 | Partly supported |
| $1B | “a billion dollars” | Moses cites 2022 revenues for BigQuery, Snowflake, and Databricks | Barr Moses | Apr 12, 2022 | Partly supported |
| 800M | “eight hundred million” | Moses cites 2022 revenues for BigQuery, Snowflake, and Databricks | Barr Moses | Apr 12, 2022 | Partly supported |
| 80% | “80%” | Moses: Automation can resolve 80% of data downtime causes | Barr Moses | Apr 12, 2022 | — |
| 20% | “20%” | Moses: Automation can resolve 80% of data downtime causes | Barr Moses | Apr 12, 2022 | — |
| 87% | “87%” | Mooney: Forbes published that 87% of data science models fail to reach production | Carolyn Mooney | Apr 12, 2022 | Supported |
| 50M | “fifty million” | Collibra raised a $250 million Series G at a $5.25 billion valuation | Matt Turck | Mar 17, 2022 | Supported |
| 50% | “50%” | Azhar: Exponential technology platforms improve 10% to 50% annually on price-performance | Azeem Azhar | Mar 8, 2022 | Supported |
| 2B | “two billion” | Eifrem: Neo4j's 2021 funding round valued the company over $2B | Emil Eifrem | Feb 7, 2022 | Supported |
| 100M | “a hundred million” | Eifrem: Neo4j crossed $100M ARR in 2021 | Emil Eifrem | Feb 7, 2022 | Supported |
| 100M | “a hundred million” | Eifrem: Only 5 modern NoSQL database companies have crossed $100M ARR | Emil Eifrem | Feb 7, 2022 | Partly supported |
| 1K | “a thousand” | Eifrem: Neo4j is frequently 1,000 times faster for connected data queries | Emil Eifrem | Feb 7, 2022 | Supported |
| 50% | “50%” | Eifrem: Neo4j is frequently 1,000 times faster for connected data queries | Emil Eifrem | Feb 7, 2022 | Supported |
| 100% | “hundred percent” | Eifrem: Neo4j is frequently 1,000 times faster for connected data queries | Emil Eifrem | Feb 7, 2022 | Supported |
| 98% | “98%” | Eifrem: GQL standard query language is 98% identical to Cypher | Emil Eifrem | Feb 7, 2022 | Partly supported |
| 100B | “a hundred billion” | Eifrem: Graph databases could become a $20B to $40B market | Emil Eifrem | Feb 7, 2022 | — |
| 40B | “forty billion” | Eifrem: Graph databases could become a $20B to $40B market | Emil Eifrem | Feb 7, 2022 | — |
| 75% | “75%” | Eifrem: Over 75% of Fortune 100 companies use Neo4j | Emil Eifrem | Feb 7, 2022 | — |
| 99% | “99%” | Eifrem: Neo4j powers 99% of all flight ticket route calculations | Emil Eifrem | Feb 7, 2022 | — |
| 85% | “85%” | Emil Eifrem: Over 85% of Neo4j's ARR originates with individual practitioners | Emil Eifrem | Feb 7, 2022 | — |
| 30% | “30%” | Numerai fell only 1.5% during the March 2020 market crash | Richard Craib | Jan 31, 2022 | — |
| 0.5% | “half percent” | Numerai fell only 1.5% during the March 2020 market crash | Richard Craib | Jan 31, 2022 | — |
| $1M | “a million dollars” | Craib: Numerai spends over $1M per year buying data | Richard Craib | Jan 31, 2022 | — |
| $10M | “ten million dollars” | Craib: Numerai paid $10 million to data scientists in 2021 | Richard Craib | Jan 31, 2022 | Supported |
| $1M | “a million dollars” | Craib: Some individual Numerai contributors have earned over $1 million | Richard Craib | Jan 31, 2022 | Not publicly verifiable |
| $15M | “fifteen million dollars” | Craib: Numerai currently has over $15 million in crypto staked | Richard Craib | Jan 31, 2022 | Supported |
| $100M | “a hundred million dollars” | Craib: Numerai charges institutional investors standard 2 and 20 fees | Richard Craib | Jan 31, 2022 | — |
| 52% | “52%” | Craib: A 0.5% accuracy gain in quantitative finance dramatically transforms Sharpe ratio | Richard Craib | Jan 31, 2022 | — |
| 0.5% | “half percent” | Craib: A 0.5% accuracy gain in quantitative finance dramatically transforms Sharpe ratio | Richard Craib | Jan 31, 2022 | — |
| 1K | “A thousand” | Craib: Numerai trades around 1,000 stocks with 500 long, 500 short | Richard Craib | Jan 31, 2022 | — |
| 1K | “a thousand” | Craib: 55% success on 1,000 assets proves edge better than Bitcoin gains | Richard Craib | Jan 31, 2022 | — |
| 55% | “55%” | Craib: 55% success on 1,000 assets proves edge better than Bitcoin gains | Richard Craib | Jan 31, 2022 | — |
| 1B | “a billion” | Katz: eBay ingests one billion events per minute into ClickHouse | Aaron Katz | Nov 16, 2021 | Supported |
| 30B | “thirty billion” | Katz: ClickHouse handled 30 billion daily events when built at Yandex | Aaron Katz | Nov 16, 2021 | Partly supported |
| 2.5B | “a half a billion” | Katz: Elastic hit $500M+ ARR with 41% international revenue mix | Aaron Katz | Nov 16, 2021 | Supported |
| 40% | “40%” | Katz: Elastic hit $500M+ ARR with 41% international revenue mix | Aaron Katz | Nov 16, 2021 | Supported |
| 41% | “41%” | Katz: Elastic hit $500M+ ARR with 41% international revenue mix | Aaron Katz | Nov 16, 2021 | Supported |
| 60% | “60%” | Meroxa's workforce is 60% Black/Brown, 40% women, and 90% underrepresented executives | DeVaris Brown | Jun 21, 2021 | — |
| 40% | “40%” | Meroxa's workforce is 60% Black/Brown, 40% women, and 90% underrepresented executives | DeVaris Brown | Jun 21, 2021 | — |
| 90% | “90%” | Meroxa's workforce is 60% Black/Brown, 40% women, and 90% underrepresented executives | DeVaris Brown | Jun 21, 2021 | — |
| 1M | “one million” | Synthesia has generated over one million videos across 40 countries | Synthesia AI Avatar | May 24, 2021 | — |
| 80% | “80%” | Riparbelli: 80% of Synthesia's enterprise clients use custom avatars | Victor Riparbelli | May 24, 2021 | — |
| 70% | “70%” | Ghodsi: Snowflake coexists with Databricks in about 70% of accounts | Ali Ghodsi | May 24, 2021 | — |
| $700M | “seven hundred million dollars” | MongoDB grew from $30M to $700M run rate under Dev Ittycheria | Dev Ittycheria | Apr 27, 2021 | — |
| 450M | “a three hundred fifty million” | MongoDB Atlas hit $350M run rate with 66 percent annualized growth | Dev Ittycheria | Apr 27, 2021 | Supported |
| 66% | “66%” | MongoDB Atlas hit $350M run rate with 66 percent annualized growth | Dev Ittycheria | Apr 27, 2021 | Supported |
| 700M | “seven hundred million” | MongoDB spent over $700M on R&D, dedicating half to free software | Dev Ittycheria | Apr 27, 2021 | — |
| 50% | “50%” | MongoDB spent over $700M on R&D, dedicating half to free software | Dev Ittycheria | Apr 27, 2021 | — |
| 99% | “99%” | MongoDB achieved 99 percent of its growth organically without acquisitions | Dev Ittycheria | Apr 27, 2021 | Not publicly verifiable |
| 1% | “one percent” | MongoDB holds just 1 percent of the massive enterprise database market | Dev Ittycheria | Apr 27, 2021 | Supported |
| 1K | “a thousand” | Pinterest runs about 1,000 parallel A/B experiments at any time | Dave Burgess | Apr 5, 2021 | — |
| 95% | “95%” | 95% of machine learning models at Google and Facebook are neural nets | Bindu Reddy | Apr 5, 2021 | — |
| 99.9% | “99.9%” | Narayan: Materialize provides 99.9% of batch database functionality | Arjun Narayan | Mar 15, 2021 | — |
| 60B | “sixty billion” | Hanlon: Reddit ingested 55-60B daily events managed by one person in 2019 | Jack Hanlon | Mar 15, 2021 | — |
| 200% | “200%” | Podjarny: Snyk grew revenue and ARR by around 200% | Guy Podjarny | Feb 17, 2021 | — |
| $50M | “fifty million dollars” | Podjarny: Snyk raised roughly $350M over 12 to 18 months | Guy Podjarny | Feb 17, 2021 | Supported |
| 50% | “50%” | Podjarny: Security products require broad stack coverage to deliver real value | Guy Podjarny | Feb 17, 2021 | — |
| 80% | “80%” | Podjarny: Security products require broad stack coverage to deliver real value | Guy Podjarny | Feb 17, 2021 | — |
| 90% | “90%” | Podjarny: Security products require broad stack coverage to deliver real value | Guy Podjarny | Feb 17, 2021 | — |
| 90% | “90%” | 90% of contacted open source projects want to contribute to OpenLineage | Julien Le Dem | Feb 1, 2021 | — |
| 90% | “90%” | Over 90% of Python tabular data passes through Pandas | Wes McKinney | Feb 1, 2021 | — |
| 50% | “50%” | Handy: Data analysts spend 50% of their time blocked by engineering | Tristan Handy | Dec 11, 2020 | — |
| 10× | “10 X” | Handy: Reverse ETL will automate workflows and multiply the data market opportunity | Tristan Handy | Dec 11, 2020 | — |
| 10× | “10 X” | Handy: Software engineers pushing to production grew over 10x in a decade | Tristan Handy | Dec 11, 2020 | Contradicted |
| 50B | “fifty billion” | Sisense's knowledge graph leverages 650 billion metadata points collected over ten years | Ashley Kramer | Nov 1, 2020 | — |
| 400% | “400%” | An airport cart client achieved 400 percent ROI within three months | Ashley Kramer | Nov 1, 2020 | — |
| 1.3B | “1.3 billion” | Bendov: ClickSoftware was acquired by Salesforce for $1.3 billion | Amit Bendov | Nov 1, 2020 | Partly supported |
| 12% | “12%” | Gong has 100 total R&D employees, including about a dozen researchers | Amit Bendov | Nov 1, 2020 | — |
| 65M | “sixty five million” | Gong had not spent a dollar of its $65M Sequoia Series C | Amit Bendov | Nov 1, 2020 | — |
| 1M | “a million” | Cancel: Enterprise software will move away from dozens of loosely coupled tools | David Cancel | Sep 25, 2020 | — |
| 50% | “50%” | Cancel: Peloton sells 40% to 50% of bikes through website messaging | David Cancel | Sep 25, 2020 | — |
| 360M | “three hundred sixty million” | Celonis bootstrapped without outside capital for its first five years | Alex Rinke | Aug 13, 2020 | — |
| 67M | “sixty-seven million” | Celonis bootstrapped without outside capital for its first five years | Alex Rinke | Aug 13, 2020 | — |
| 80% | “80%” | Celonis's first customer cut ticket resolution from five days to same-day | Alex Rinke | Aug 13, 2020 | — |
| 9B | “nine billion” | Pesenti in 2020: 9B parameter Blenderbot is the largest chat model | Jerome Pesenti | Jun 10, 2020 | Partly supported |
| 10× | “10 X” | Pesenti in 2020: 10x annual AI compute scaling is economically unsustainable | Jerome Pesenti | Jun 10, 2020 | Held up |
| 52% | “52%” | Horowitz: a16z is 52% female with seven female general partners | Ben Horowitz | Feb 20, 2020 | — |
| 1T | “a trillion” | Cerebras's WSE is the first chip with over 1 trillion transistors | Michael James | Feb 20, 2020 | Supported |
| 10,000× | “10,000 times” | Cerebras's chip provides 10,000x the memory bandwidth of standard GPUs | Michael James | Feb 20, 2020 | Supported |
| 13× | “13 x” | Google Anthos drove a 13x deployment velocity gain for a financial client | Amr Awadallah | Feb 20, 2020 | — |
| $10B | “ten billion dollars” | Google Cloud reached a $10B annual revenue run rate by early 2020 | Amr Awadallah | Feb 20, 2020 | Supported |
| $700M | “seven hundred million dollars” | Cloudera reached a $700M annual revenue run rate by late 2019 | Amr Awadallah | Feb 20, 2020 | Supported |
| 10T | “10 trillion” | Pomel: Datadog processes about 10 trillion records per day | Olivier Pomel | Jan 22, 2020 | — |
| $20M | “twenty million dollars” | Delangue: Hugging Face raised over $20M from Lux, Betaworks, and Kevin Durant | Clement Delangue | Jan 22, 2020 | — |
| 1K | “a thousand” | Delangue: Over 1,000 companies including Apple and Bing use Hugging Face | Clement Delangue | Jan 22, 2020 | — |
| 1% | “a percent” | A fractional drop in AI performance can cost hundreds of millions | Adam Wenchel | Jan 22, 2020 | — |
| 10% | “10%” | Enterprise AI adoption fails without risk mitigation despite performance gains | Adam Wenchel | Jan 22, 2020 | — |
| $1M | “a million dollars” | Elsamadisi: WeWork's 45-person data team spent over $1M with little impact | Ahmed Elsamadisi | Nov 13, 2019 | — |
| 3× | “three times” | Anthony: Pedestrians physically punched Cruise autonomous cars in 2017 | Sam Anthony | Nov 13, 2019 | Contradicted |
| 30M | “thirty million” | Unacast has raised $30 million with engineering team in Oslo | Thomas Walle | Nov 13, 2019 | — |
| 10% | “10%” | Walle: Unacast's anonymized dataset covers roughly 10% of the US population | Thomas Walle | Nov 13, 2019 | — |
| 5× | “five times” | Volpi: Index Ventures made 5x return on Hortonworks investment | Mike Volpi | Nov 13, 2019 | — |
| 95% | “95%” | Volpi: Employees write 95% of code in open source startups | Mike Volpi | Nov 13, 2019 | Supported |
| 5% | “five percent” | Volpi: Employees write 95% of code in open source startups | Mike Volpi | Nov 13, 2019 | Supported |
| 90% | “90%” | Volpi: AWS holds an estimated 85% to 90% market share among developers | Mike Volpi | Nov 13, 2019 | — |
| 20% | “20%” | Volpi: Index Ventures' portfolio has roughly 20% female CEOs | Mike Volpi | Nov 13, 2019 | — |
| $100M | “a hundred million dollars” | Tray.io had raised over $100 million in funding by October 2019 | Rich Waldron | Oct 22, 2019 | Supported |
| 2.2× | “2.2 x” | Companies with automated sales pipelines grow 2.2x faster than competitors | Rich Waldron | Oct 22, 2019 | Contradicted |
| 7.4B | “7.4 billion” | CTRL-Labs uniquely identifies users among 7.4 billion in 800 milliseconds | Thomas Reardon | Sep 17, 2019 | — |
| $10B | “ten billion dollars” | Guo: Greylock maintains a single fund rather than multi-billion opportunity vehicles | Sarah Guo | Sep 17, 2019 | — |
| 80B | “eighty billion” | Guo estimates the 2019 cybersecurity market at $75B to $80B | Sarah Guo | Sep 17, 2019 | Contradicted |
| $40B | “forty billion dollars” | Guo: Most of 2019's $40B security software spend remains on-premise | Sarah Guo | Sep 17, 2019 | Partly supported |
| 100× | “hundred X” | Yang: Apache Druid can condense raw data 100x through roll-up aggregation | FJ Yang | Jun 12, 2019 | — |
| 5% | “five percent” | Benn Stancil: Data teams average 2% to 5% of tech headcount | Benn Stancil | May 13, 2019 | — |
| 4% | “four percent” | Benn Stancil: Data teams average 2% to 5% of tech headcount | Benn Stancil | May 13, 2019 | — |
| 100% | “hundred percent” | Text IQ achieved profitability by 2017 with a 100% pilot conversion rate | Apoorv Agarwal | May 13, 2019 | — |
| $4M | “four million dollars” | Legacy compliance review for a healthcare client would have cost $4 million | Apoorv Agarwal | May 13, 2019 | — |
MADLEDGER All shows Try the podcast clipper →