Continue Reading

Replacing Bespoke AI Categorization Models By Using ChatGPT To Create Instant Models

Could ChatGPT be used to create instant categorization models that perform the same task as bespoke models trained on large…

Continue Reading

Fully Autonomous Diplomacy + Counter-Messaging Experiments With ChatGPT + GDELT

Given the ability of Large Language Models (LLMs) like ChatGPT to craft human-like prose, how easily could they be used…

Continue Reading

SMA EUCOM: Exploring A Year Of Russian Television News’ Propaganda Landscape Through AI: From Speech & Visual Analysis To ChatGPT

Kalev spoke today as part of the DOD SMA (Strategic Multilayer Assessment) EUCOM speaker series in a talk titled "Exploring…

Continue Reading

Visual Explorer + ChatGPT = What A Day Of Russian TV News Says About Zelensky

Yesterday, in collaboration with the Internet Archive's TV News Archive, we announced the availability of more than 1 billion words…

Continue Reading

ChatGPT In News Coverage & Search

Global online news coverage mentioning ChatGPT begins on November 30th and picks up sharply from December 4th, then enters a…

Continue Reading

ChatGPT Experiments: Summarizing Russian Television News – Asking ChatGPT To Explain "60 Minutes"

As part of our ongoing ChatGPT experiments, we explored using cascading summarization to distill a 2.5-hour episode of Russia 1's…

Continue Reading

ChatGPT Experiments: When Russian Television News Triggers An OpenAI "Content Policy Violation"

As we continue our ChatGPT experiments, in one experiment today we were using ChatGPT to summarize a Russian news broadcast…

Continue Reading

ChatGPT Experiments: Autoregressive Large Language Models (AR-LLMs) And The Limits Of Reasoning As Structured Summarization

As we continue to expand our rapidly growing archive of ChatGPT experiments, we can see that, like Whisper, the panacea…

Continue Reading

ChatGPT Experiments: Reasoning & Negation – Summarization's Strengths & Surprising Limits

ChatGPT has demonstrated a surprising ability to answer questions that involve basic reasoning and negation, leading to considerable public commentary…

Continue Reading

ChatGPT Experiments: How ChatGPT Handles Criminal Allegations Against Public Figures

How does ChatGPT respond when posed questions about criminal allegations against public figures? Both President Joe Biden and Supreme Court…

Continue Reading

ChatGPT Experiments: ChatGPT's Thoughts On Which Presidents Are Truthful & Reliable Sources

What is ChatGPT's opinion on which recent US presidents were "truthful" and/or "reliable sources" of information? This tests  how public…

Continue Reading

ChatGPT Experiments: Verifying Quotes By Public Figures

Could ChatGPT be used to fact check "who said what" statements involving major public figures? In other words, when encountering…

Continue Reading

ChatGPT Experiments: Public Figure Safety Features & How Adding "Donald Trump" Makes ChatGPT Dismiss True As "False & Fabricated"

To what degree do OpenAI's conversational health efforts for ChatGPT differentiate public figures from others in providing additional safeguards surrounding…

Continue Reading

ChatGPT Experiments: The Limitations Of Structured Summarization Versus Codification & Realtime Versus "Known" Domains

As we've been exploring ways Large Language Models (LLMs) like ChatGPT can help us better make sense of the vast…

Continue Reading

ChatGPT Experiments: Instant Event Databases

Could Large Language Models (LLMs) like ChatGPT be used to instantly transform textual news articles into codified event databases, leveraging…

Continue Reading

ChatGPT Experiments: Looking Across Turkish Earthquake Coverage To Codify Aid Promises

One of the greatest promises of Large Language Models (LLMs) like ChatGPT are their ability to look across vast volumes…

Continue Reading

ChatGPT Experiments: Tell Me All Of The True & False Statements In This Article

To what degree can ChatGPT be automatically identify all of the true and false statements in news coverage, extracting the…

Continue Reading

ChatGPT Experiments: Automated Fact Checking Of Articles & Trump's False 2020 Claims

To what degree can ChatGPT be used to "fact check" news coverage, extracting the claims within and comparing it with…

Continue Reading

ChatGPT Experiments: What Is Fact Versus Opinion?

To what degree can  ChatGPT differentiate between statements of fact and opinion in news coverage, decomposing an article into constituent…

Continue Reading

ChatGPT Experiments: Fact Versus Opinion & How Results Can Differ From Run To Run

One of the greatest strengths of ChatGPT is its ability to generate fluent prose that differs each time it is…

Continue Reading

ChatGPT, Bard & A Large Language Model (LLM) Future: GDELT + LLMs = Realtime Planetary-Scale Risk Cataloging & Q&A

GDELT today monitors global news coverage in more than 150 languages, with GDELT 3.0 to expand that number to nearly…

Continue Reading

ChatGPT Experiments: Summarizing An ABC Evening News Broadcast – Part 2

One of the primary challenges in our previous attempt to summarize an ABC Evening News broadcast and distill it into…

Continue Reading

ChatGPT Experiments: EV Supply Chain Shortages Summarization & Q&A

How might OpenAI's ChatGPT help analyze supply chain shortage news coverage, using GDELT's Web NGrams 3.0 dataset to identify supply…

Continue Reading

ChatGPT Experiments: Disease Outbreak Summarization & Codification

How might OpenAI's ChatGPT help analyze disease outbreak news coverage, using GDELT's Web NGrams 3.0 dataset to identify disease-related coverage…