Thinking on trustworthy, human-led AI

When AI gets expensive to run, guess who pays

Three AI companies said in their own words that the compute behind their products is scarce and expensive. Here's how that bill reaches customers, and the questions I'd ask before it reaches yours.

Who changed your model?

AI vendors can now swap the model doing your work without telling you. In most of life that's fine. In regulated work, it's a change you'd normally have to sign off on.

Workslop and the depth you can't fake

Generic AI output hurts small businesses most, because they can't lean on a famous name. The good news: depth still shows up in the follow-up questions, and that's where small firms win.

Know what's authoritative before the AI reasons

Before any AI reasons over a firm's data, people have to decide which sources count, what terms mean and where they conflict. Knowledge graphs make those decisions visible.

Human-in-the-lead, not human-in-the-loop

Human-in-the-loop has quietly turned into "a human approved it." A new study shows why I'd rather have a human in the lead, and what that takes.

Your data is a line item in someone's bankruptcy

When an AI company fails or is sold, the data its customers handed over becomes an asset, and the privacy promises attached to it can pass to an owner who never made them.

A great model isn't a governed system

The newest AI tool for financial advisors says it works "with the access you already have." That one line tells you where the real work is.

Opt-out theater

Researchers keep testing whether privacy opt-outs actually work, and the results aren't reassuring. The only opt-out I fully trust is deciding what never leaves the building.

Every data opportunity is a nail

Treating all data as training material is a choice someone made for you. Firms should make their own choice about what their data is for.

Cognitive Science Techniques for Personalized Treatments: New Paradigms for Precision Care

The rapid advancement of artificial intelligence (AI) and the integration of cognitive science methodologies has opened up exciting opportunities for improving healthcare outcomes.

Written February 1, 2023 · Added to reclassify.ai April 2024

Product Ontology and Semantic Engines for Master Data Management

Master Data Management (MDM) is a critical aspect of modern business operations across various industries.

Written July 1, 2022 · Added to reclassify.ai April 2024

Enhanced Threat Intelligence Automation

With the rise of sophisticated cyber threats, organizations across industries have increasingly turned towards threat intelligence as a critical element of their defense strategies.

Written April 16, 2021 · Added to reclassify.ai April 2024

Empowering Asset Management Firms with Cognitive Science Solutions

Asset management firms rely heavily on data analytics, financial modeling, risk assessment, and investment decision-making processes to generate returns for their clients.

Written December 15, 2020 · Added to reclassify.ai April 2024

Semantic Data Platforms

We often get asked what exactly we mean when we talk about creating a semantic framework for data intelligence in a firm or organization.

First published on reclassify.ai on November 9, 2020

Addressing Racial Disparities in Maternal Mortality using Cognitive Science Technology

Maternal mortality remains a significant public health challenge worldwide, particularly among women of color.

Written October 30, 2020 · Added to reclassify.ai April 2024

Knowledge Graph Sweet Spots

At Reclassify AI, we are of the firm belief that knowledge graph technology can and should form the backbone of any organization's approach to data management and governance.

First published on reclassify.ai on April 6, 2020

Smart Data and Smart Management

I have spent the last 8 years of my 17 year career working in financial services environments that are grappling with some of the most complex data challenges.

First published on reclassify.ai on August 29, 2019