Blog

Research

KL3M: The First Fairly Trained Large Language Model

KL3M shows that large language models can be built on copyright-clean training data, with provenance that enterprises can actually defend.

Read more
Engineering

CISA and FBI Call for Memory Safety Roadmaps: Is C++ on Borrowed Time?

CISA and the FBI are turning memory safety from an engineering preference into a board-level issue, and C/C++ is suddenly on the defensive.

Read more
Data Strategy

The New York Times Sues OpenAI: The Copyright Case That Could Define AI Training

The New York Times' lawsuit against OpenAI and Microsoft turns AI training data provenance, fair use, and output risk into a very expensive conversation.

Read more
AI Governance

EU AI Act Trilogue Complete: The Final Text and What It Means

The EU institutions have reached political agreement on the AI Act, and the first comprehensive AI law in the world is now moving from theory to enforcement.

Read more
AI Governance

The Bletchley Declaration: 28 Nations Agree on AI Safety (But Not on How)

The Bletchley Declaration is a real signal that AI safety has gone global, but the hard part is still turning shared concern into enforceable rules.

Read more
AI Governance

Biden's AI Executive Order: The Most Comprehensive Federal AI Action to Date

Executive Order 14110 turns AI policy into a real compliance agenda, with federal demands on safety testing, watermarking, privacy, and civil rights.

Read more
Privacy & Security

MGM Resorts Hacked: Social Engineering Still Beats Technical Controls

MGM Resorts’ September 2023 cyber incident is a reminder that the easiest path into a hardened environment is often a human with a phone and a reset button.

Read more
Engineering

HashiCorp Goes BSL: What the Terraform License Change Means for Your Infrastructure

HashiCorp's move to BSL 1.1 for Terraform and other core products changes the rules for vendors, integrators, and anyone packaging infrastructure tooling commercially.

Read more
Privacy & Security

SEC Adopts Cybersecurity Disclosure Rules: 4-Day Incident Reporting Begins

The SEC's new cyber rules turn incident response into a filing deadline problem, and the 4-business-day clock starts at materiality.

Read more
AI Governance

The EU AI Act Passes Parliament: What US Companies Should Start Preparing Now

The European Parliament's June 14 vote signals that AI governance is moving from theory to enforcement, and U.S. companies need to get their house in order now.

Read more
Privacy & Security

Meta's EUR 1.2 Billion Fine: The End of Unchecked Transatlantic Data Flows

The Irish DPC's record fine against Meta is a reminder that cross-border data flows need more than SCCs, more than supplementary measures, and definitely more than hope.

Read more
AI Governance

The Samsung ChatGPT Leak: Why Every Company Needs an AI Acceptable Use Policy

Samsung’s ChatGPT leak is a blunt reminder that without an AI acceptable use policy, employees will paste confidential data into whatever tool seems helpful.

Read more