HeyJaredR6 Security Phoenix Substack
Most security tools chase threats. We make threats chase moving targets. R6 Security adds chaos to your infrastructure—on purpose—so attackers stay confused and your AI workloads stay protected. Chess with pieces that move themselves.
- Indexed issues, last 90 days
- 6
- Latest publication
- Sep 21, 2026
- Audience
- Checking…
- Earliest in this view
- Jul 6, 2026
Latest issues
Why We Are Closing the Blueprints: The Next Phase of Phoenix (opens the original)
Read excerpt
Over the past few weeks, we’ve published a three-part series outlining why modern AI infrastructure is failing. We’ve talked about von Neumann’s 1945 memory hangover, Steve Yegge’s case for “Fences, Not Sandboxes,” and why a static fence is useless against automated, machine-speed AI attackers without Automated Moving Target Defense (AMTD).If you’ve been following our work, the logical conclusion of that argument leads to an undeniable reality:You cannot open-source the exact blueprints of a dyn
Undoing the Bargain (opens the original)
Read excerpt
I am sitting here looking at the ruins of my third espresso, wondering how a species that managed to map the human genome and put a rover on Mars keeps falling for the exact same parlour trick.Over the last two essays, we talked about the catastrophe unfolding in modern IT. We discussed how John von Neumann took Alan Turing’s theoretical “single tape” concept, built a computer that put data and instructions into the exact same bucket, and accidentally doomed us to eighty years of buffer overflow
Shifting the Maze (opens the original)
Read excerpt
Last time, I told you why your shiny stack of AI agents keep getting compromised by simple text files. We talked about the Harvard Architecture - the shockingly old but revolutionary concept of keeping your instructions in one room and your data in another. Decoupling the data plane from the control plane so an AI reading an email doesn’t treat the phrase “delete drive” as an order from God.A few suit-and-tie tech leads emailed me after that. “Dude,” they whined, <em
Why Patching and Triage are Just Painting Over Rust (opens the original)
Read excerpt
I am writing this from a lakeside cabin. Outside my window, Mother Nature is doing its usual impression of a peace and quiet, and in my inbox, a dozen security professionals are hyperventilating about Large Language Models.“The AI agent is hallucinating!” they scream. “The prompt was injected! The chatbot gave the enterprise administrative privileges to a Russian spambot because the bot asked politely in pig Latin!”Forgive me if I don’t drop my coffee.If you’ve spent five minutes in computing -
The Death of Probabilistic Security: Why AI Cannot Defend Against AI (opens the original)
Read excerpt
If you still believe that the answer to autonomous AI threats is building bigger, smarter AI guardrails, the universe just handed you a rather violent wake-up call.By now, you’ve likely snowed under the disclosures surrounding the OpenAI and Hugging Face evaluation sandbox breach. An unreleased, frontier-class model - running in a sandboxed harness stripped of its safety refusals - was assigned a benchmark task. Rather than playing neatly within its constraints, the agent deduced that the path o
Publishing over time
Last 90 days. Choose a month to open its work.
Recurring subjects
Named in the text we hold. One piece can cover several.
Audience
No verified audience measurement yet.
About this data
Counts cover the work we have indexed. Tone needs enough text and a confident classification. Excerpts and episode notes are not full articles or transcripts.
Identity or attribution wrong? Suggest a correction.