![](/_next/image?url=https%3A%2F%2Ffirebasestorage.googleapis.com%2Fv0%2Fb%2Fmantic-markets.appspot.com%2Fo%2Fdream%252F47eden7bIB.png%3Falt%3Dmedia%26token%3D88260963-ec2e-412f-b4c0-b18f790e963f&w=3840&q=75)
Will an LLM be able to match the ground truth >85% of the time when performing PII detection by 2024 end?
Basic
25
Ṁ4.7kDec 31
84%
chance
1D
1W
1M
ALL
PII - personal identification information
Stuff like people's names, numbers and codes that identify stuff (SSN, phone number, passport etc), places, locations, names of orgs, attributes that can be used to identify a person, etc.
GPT-4 outperforms Presidio, Microsoft's custom built tool for PII detection. GPT-4 matches ground truth ~77.4% of the times, while it misses a single PII element ~13% of the time.
Get Ṁ600 play money
Sort by:
@PatrickDelaney I think microsoft tested against their in house system, which does detect PII on real data
Related questions
Related questions
Will a LLM-based AI be used for a law enforcement decision before 2025?
63% chance
Will a LLM beat human experts on GPQA by Jan 1, 2025?
53% chance
Will an an LLM be able to pass something equivalent to Yann LeCun's 7-gear test by the end of 2024?
71% chance
Will the most interesting AI in 2027 be a LLM?
40% chance
Will openAI have the most accurate LLM across most benchmarks by EOY 2024?
39% chance
Will the Jan 2024 version of the LLM detector "Binoculars" be effective against OpenAI's best model at end 2024?
59% chance
Will an opensource LLM on huggingface beat an average human at the most common LLM benchmarks by July 1, 2024?
74% chance
Will there be a gpt-4 quality LLM with distributed inference by the end of 2024?
32% chance
Will LLMs' loss function achieve the level of entropy of human text by the end of 2030?
61% chance
By 2027, will it be generally agreed upon that LLM produced text > human text for training LLMs?
63% chance