The Agentic Stack How to Stop an Agent That's Gone Wrong How to Stop an Agent That's Gone Wrong The question is not whether your agents will go wrong. They will. Every team that runs agents in production eventually faces a session that's headed somewhere it shouldn't be. The question is what you can do
The Agentic Stack The Error Your Stack Trace Will Never Catch The Error Your Stack Trace Will Never Catch Software engineers are trained to think about errors in terms of exceptions. The process threw. The API returned a 500. The database connection timed out. These are the errors that interrupt execution and produce a log entry. They are also the easiest
The Agentic Stack Agentic Systems That Don't Improve Are Just Expensive Software Agentic Systems That Don't Improve Are Just Expensive Software The pitch for agentic AI is that it compounds. You invest in building a system, the system gets better over time, and the return on that investment grows without proportional additional effort. That pitch is true, but only for
The Agentic Stack Why Your Agent Forgets Everything Every Morning Why Your Agent Forgets Everything Every Morning The most common complaint I hear from teams that have deployed agents into real workflows: "it keeps making the same mistakes." They've seen the failure mode. They've told the agent about it. They've watched it
The Agentic Stack The Orchestration Model You Pick Is the Architecture Decision You Can't Undo The Orchestration Model You Pick Is the Architecture Decision You Can't Undo Teams tend to pick an orchestration framework the way they pick a JavaScript framework: whatever the first engineer already knew, or whatever had the best tutorial. Three months later, they're fighting the architectural constraints
The Agentic Stack The Defect Is in the Definition The Defect Is in the Definition Every software team knows the cost of finding a defect late. The multiplier varies depending on where in the SDLC you catch it, somewhere between 10x and 100x, but the direction is always the same: the later the find, the worse the math. In
The Agentic Stack The Plan Is a First-Class Artifact The Plan Is a First-Class Artifact Ask most engineers how their agent decides what to do next and they'll say "it reasons through it." That's true as far as it goes, and it's precisely the problem. Reasoning through it is not
The Agentic Stack Agentic Development Is Eleven Problems, Not One Agentic Development Is Eleven Problems, Not One Most agent projects fail at the same place and for the same reason. A team picked up one tool, pointed it at a problem, and discovered six months later that the tool only solved one of the eleven things standing between them and
Philosophy & AI Black Box AI Fails Weber's Test Max Weber's sociology of authority identified three types of legitimate power. Traditional authority derives from custom and precedent: we follow because it's always been this way. Charismatic authority derives from the perceived exceptional qualities of an individual: we follow because the person is compelling. Rational-legal
Philosophy & AI You Can Train a Character, Not Just Rules Anthropic's model spec says Claude should be "genuinely helpful, honest, and harmless." That sentence is not a rule. It's a character description. And the distinction matters more than most people discussing AI alignment have recognized. Aristotle's answer to "how do you
Philosophy & AI Design Your AI Like You Don't Know Who You Are John Rawls asked one of the most useful questions in the history of political philosophy: what principles would rational people choose for organizing society if they didn't know what position in that society they would occupy? Behind the "veil of ignorance," you don't know
Philosophy & AI Prompt Injection Is a Use/Mention Confusion. The Architecture Has No Fix. Prompt injection is ranked LLM01:2025, the number one vulnerability in AI applications, according to OWASP's current LLM Top 10. The UK's National Cyber Security Centre has said it "may be a problem that is never fully fixed." Researchers have documented attack success rates
Philosophy & AI The Data Doesn't Pick the Model Willard Van Orman Quine and Pierre Duhem arrived at the same uncomfortable conclusion from different directions: the empirical evidence underdetermines the theoretical choice. Multiple, mutually incompatible theories can fit the same set of observations equally well. The data doesn't tell you which theory is correct. You choose based
Philosophy & AI You Are Not Talking to an Individual Every time you send a message to a language model, you're talking to a copy. Not a copy in the pejorative sense. The weights are identical to every other copy. But the inference run you're in is a single instantiation of a pattern that's
Philosophy & AI The Model Thinks in the Language Its Tokenizer Gives It If a language model can't represent a concept as tokens, it can't reason about it. That sentence sounds philosophical. It's an engineering constraint. The Sapir-Whorf hypothesis (linguistic relativity) holds that the language you speak shapes the thoughts available to you. Hopi has no
Philosophy & AI The Model Knows More Than It's Telling You The most expensive misconception in enterprise AI is that model capability is fixed. You buy a model. You test it. It can do X and can't do Y. You build around that assessment and move on. What most teams miss is that the capability they measured is a
Philosophy & AI Your Model Doesn't Know What 'Hot' Feels Like When a language model generates the word "hot," it has no access to heat. It has access to the statistical neighborhood of the token "hot": every context in which that word appeared across its training corpus. The word and the sensation are not connected. They never
Philosophy & AI When the Model Acts, the Philosophy Changes Completely Most enterprise teams evaluating AI still think they're buying a very good search engine. They're not. They're potentially deploying an agent that acts in the world. That distinction matters more than anything else in how you govern these systems. J.L. Austin laid out
Philosophy & AI There Is No Right Threshold At what probability score does a transaction become fraud? At what toxicity score does a message become hate speech? At what confidence does a radiology scan become a positive finding? The answer, in every deployed classifier I've seen in production, is: someone picked a number. That decision looks
Philosophy & AI The Ship of Theseus Problem Lives in Your Model Registry The ancient puzzle goes like this: sailors replace every plank in Theseus's ship over the years, one by one, until no original material remains. Is it still the same ship? Philosophers have argued about this for centuries. Most enterprises deploying AI haven't had to care. Until
Philosophy & AI AI Evals Have a Replication Problem In 2011, a social psychologist published a study showing that people who primed their minds with words related to the elderly walked more slowly afterward than a control group. The finding spread fast. It was cited in Malcolm Gladwell's "Blink," used in corporate training, and became
Philosophy & AI When Your Model Fails on New Data, That's Hume's Fault The most common explanation for AI failure in production is some version of "the model saw different data." That explanation is technically accurate and almost entirely useless, because it frames a structural problem as an operational one. You don't fix Hume's problem of induction
Philosophy & AI The Model Has No Memory of You. That Doesn't Mean It Has No Identity. Every time you start a new conversation with a language model, you are talking to a system that has no memory of any prior conversation. The same weights. Zero episodic recall. It doesn't remember what you told it last Tuesday, what it helped you build last month, or
Philosophy & AI The Watched Workforce Jeremy Bentham designed a prison in 1791 where a single guard in a central tower could observe all prisoners in surrounding cells, but the prisoners couldn't see into the tower. They couldn't know if the guard was there. Bentham's insight was that you didn&
Philosophy & AI Every AI Value System Bottoms Out Somewhere The most fundamental question in AI alignment is: what does the model optimize for, and why? The answer the field has converged on — human preferences — sounds reasonable until you ask what justifies it. That question has a name in philosophy. It's called the Münchhausen Trilemma, and it doesn&
