[Go to site: main page, start]
TechTalks

TechTalks

Home
Archive
About
Scaling the harness: The next major bottleneck in agentic AI
Scaling LLMs hits limits when dealing with agentic AI tasks. For that, we need to look at the harness and the system built around the model(s).
 • Ben Dickson
What makes Cursor's Composer 2.5 a good coding model (and what are the caveats)
A deep look at the self-distillation techniques that make Composer 2.5 such a great coding model (and the hidden tradeoffs they introduce to AI…
 • Ben Dickson
How prompt injection broke Nvidia's sandboxed OpenClaw agent
Research into Nvidia’s NemoClaw reveals that sandboxes don't stop AI agents like OpenClaw from leaking data. We need to rethink security from first…
 • Ben Dickson
Unpacking Gemma 4’s multi-token prediction (and why you should care)
How Gemma 4’s multi-token prediction and community-driven DFlash are speeding up local LLM throughput by 3-6x.
 • Ben Dickson
How to scale LLMs to 100 million tokens without blowing up memory costs
Memory Sparse Attention (MSA) scales LLM context windows to an unprecedented 100 million tokens while preserving accuracy.
 • Ben Dickson
The silent threat of Claude Code (and other AI coding assistants)
A new study reveals how AI coding assistants like Claude Code are quietly hoarding and publishing sensitive API keys to code repositories.
 • Ben Dickson
The 'by design' security flaw of Model Context Protocol (MCP)
Security researchers have uncovered a massive architectural flaw in Anthropic's Model Context Protocol, exposing millions of AI applications to remote…
 • Ben Dickson
The hidden trap of LLMs self-distillation
Optimizing LLMs for concise answers can destroy their ability to explore alternative solutions on difficult problems. New study reveals the hidden cost…
 • Ben Dickson
The art of AI harness engineering
The recent leak of Anthropic's Claude Code reveals a hard truth: as LLMs become commoditized, the sophisticated engineering harness built around them is…
 • Ben Dickson
How GhostClaw exploits macOS and OpenClaw to steal developer credentials
As developers rush to run local AI agents on Mac Minis, GhostClaw malware exploits macOS binaries to silently harvest credentials.
 • Ben Dickson
Inside V-JEPA 2.1, the huge upgrade to Meta's world model
AI models have historically struggled to balance motion tracking with spatial detail. Meta’s V-JEPA 2.1 solves this, pushing the boundaries of video…
 • Ben Dickson
AI won't kill SaaS, but major shifts are coming
The recent tech selloff sparked fears of a SaaSpocalypse. Here is why the death of software subscriptions is a myth, and how AI agents are creating a…
 • Ben Dickson
TechTalks
TechTalks
In-depth discussions about machine learning, deep learning, reinforcement learning, neural networks, artificial general intelligence, AI business, and other technology trends.
Recommendations
View all 15
NEW ECONOMIES
NEW ECONOMIES
Ollie Forsyth
22nd Century Frontier™
22nd Century Frontier™
Petar Dimov
AI Side Notes with Sharon Goldman
AI Side Notes with Sharon Goldman
Sharon Goldman
The Business Engineer
The Business Engineer
Gennaro Cuofano
The VC Corner
The VC Corner
Ruben Dominguez
© 2026 Ben Dickson · Privacy ∙ Terms ∙ Collection notice
Start your SubstackGet the app
Substack is the home for great culture
    This site requires JavaScript to run correctly. Please turn on JavaScript or unblock scripts