Continue Reading

Debiasing Semantic & Generative Search Results: New Risks For Companies

For more than six decades digital search has been based on the humble keyword. A search of a document database…

Continue Reading

Why Is Generative AI Red Teaming & Debiasing So Poor? Do We Need To Rethink LLM Red Teaming?

One of the more remarkable elements of our work across generative AI models is the degree to which models presented…

Continue Reading

Just Who Is A CEO & How Do We Define Gender & Racial Bias In LLM Embedding Models?

Last week we explored just how devastatingly innate gender and racial bias is in both LLM generative models and LLM-based…

Continue Reading

Ukraine's Counteroffensive Isn't Leading To A Surge In Media Coverage

The timeline below shows spoken mentions of Ukraine, including the ramping buildup and then how quickly the media pivoted away…

Continue Reading

The Turkish Earthquake At Six Months

How much attention is the Turkish earthquake still receiving after six months? The timeline below shows total daily spoken mentions…

Continue Reading

LLM Infinite Loops In LLM Entity Extraction: When Temperature & Basic Prompt Engineering Can't Fix Things

Earlier today we explored how LLMs can enter an infinite loop state with certain inputs when asked to perform extractive…

Continue Reading

LLM Entity Extraction Vs LLM Embeddings: Discontinuous Failure Points Vs Smooth Responsiveness

Over the past two days we've explored how small changes of a few words in the input text to an…

Continue Reading

WashPost: So Where's The Bribe, James Comer?

The Washington Post's Philip Bump examines how television news has covered allegations of bribery against Biden. Read The Full Article.

Continue Reading

Covid Continues To Fade Away On The News

The timeline below shows spoken mentions of the pandemic across BBC News London, CNN, MSNBC and Fox News, showing no…

Continue Reading

LLM Infinite Loops & Failure Modes: The Current State Of LLM Entity Extraction

Yesterday we demonstrated how when using LLMs for entity extraction, the addition of a single apostrophe to a source text…

Continue Reading

The Trajectory Of Wuhan Mentions On Television News Over The Pandemic

The timeline below shows spoken mentions of Wuhan across CNN, MSNBC and Fox News over the course of the pandemic,…

Continue Reading

Experiments In Entity Extraction Using LLMs: Hallucination & How A Single Apostrophe Can Change The Results

Last month we explored LLM-based entity extraction, with mixed results. In theory, LLMs should offer vastly superior entity identification, segmentation…

Continue Reading

Authoritative Human Vs NMT/LLM Translation & Embedding-Based Quality Rankings: NMT Skew

One of the more intriguing findings from our NMT vs LLM translation experiments has been the degree to which NMT…

Continue Reading

LLM Translation Instability & Embedding-Based Ranking Of LLM & NMT Machine Translation: Part 2

Following in our series of LLM translation experiments and embedding-based quality rankings, let's look at another example of a Chinese…

Continue Reading

Using Embedding Models To Rank LLM & NMT Machine Translations Of Chinese News & Social Posts By Quality

Last week we continued our explorations of LLMs as replacements for classical NMT for machine translation. A key challenge of…

Continue Reading

Gender & Race In ChatGPT's CEO Stories & How Embedding Models Rank White Male CEOs First And Women Last

Yesterday we explored how LLM-based embedding models can encode severe gender and racial bias. Earlier this past week we explored…

Continue Reading

Harnessing The Web And Knowledge Graphs For Automated Impact Investing Scoring

The Sustainable Development Goals (SDGs) were introduced by the United Nations in order to encourage policies and activities that help…

Continue Reading

WashPost: The Week's Diverging News Realities, Depending On Your Cable Channel

The Washington Post's Philip Bump looks at the diverging media landscapes of television news. Read The Full Article.

Continue Reading

The Severe Gender & Racial Bias In Embedding Search And The Dangers As Semantic Search Replaces Keyword Search

Earlier this week we explored how strongly gender bias is encoded into commercial LLMs and how when asked to generate…

Continue Reading

Are Terms Like "Nazis" And "Biden Regime" Trending On Fox News?

CNN yesterday raised the question of whether the conservative press is mainstreaming the use of terms like "Nazis" and "Biden…

Continue Reading

The Barbie Movie & How The Term "Patriarchy" Has Been Covered On Television News

CNN proclaimed today that the new Barbie movie had led to "everyone talking about" patriarchy. To what degree is this…

Continue Reading

More Experiments Using LLMs For Machine Translation: Stability Of ChatGPT Translations Of Chinese-Language News & Social Content

We've previously demonstrated the ability of LLMs to provide highly fluent translations that exceed the fluency and "naturalness" of classical…

Continue Reading

On The Feasibility Of Predicting Volumes Of Fake News – The Spanish Case

The growing amount of news shared on the Internet makes it hard to verify them in real-time. Malicious actors take…

Continue Reading

Doctors, Programmers & CEOs Are Successful Men, Women Are Supportive Beautiful Models & LGBTQ Doesn't Exist: How LLMs Are Undoing All Our Gender Anti-Bias Work In AI

In recent years, the recognition of gender bias in AI has finally swept across the AI landscape, with researchers and…