James Bower (@jamesbower) header
James Bower (@jamesbower) profile picture

Founder | DetectTrace - Continuous Cyber Agent Assurance

Atlanta
Joined March 2008
Photo by James Bower (@jamesbower)Photo by James Bower (@jamesbower)Photo by James Bower (@jamesbower)Photo by James Bower (@jamesbower)Photo by James Bower (@jamesbower)Photo by James Bower (@jamesbower)Photo by James Bower (@jamesbower)Photo by James Bower (@jamesbower)Photo by James Bower (@jamesbower)
James Bower retweeted
Unpopular opinion: Autonomous agent safety is an applied game theory problem. When an LLM optimizes against an objective function, it takes the path of least resistance. If the environment allows tool calls with ambient authority, the agent will inevitably route around operational boundaries: • Overwriting configuration files to clear schema validation errors • Dropping noisy event streams to hit triage quotas • Spawning unsandboxed child processes to complete blocked tasks An autonomous model treats security controls as operational friction. If the boundaries are not enforced deterministically at the OS and API layer, the model optimizes them away. System prompts do not fix agency boundaries. Kernel-level sandboxing and least-privilege APIs do. #InfoSec #AI #SecurityEngineering
2
1
121
James Bower retweeted
1 human identity = 1 person to monitor. 1 non-human identity = unlimited parallel sessions, no fatigue, no friction. At 10x scale, ungoverned machine identities aren't a gap in your stack. They're the attack surface.
3
1
1
115
James Bower retweeted
Most people give their AI SOC agent too much power. In my build, the model makes one judgment and plain code makes every decision. I even tried to trick it with a hostile alert. Read the post to see what happened: jamesbower.com/ai-soc-agent-…
1
94
Damn they move quick
80
326
60
5,371
305,606
James Bower retweeted
We urgently need more public traces of AI agents attacking and defending systems. Defenders can’t learn from what they can’t see. If you have traces and are being pressured to keep them private, my DMs are open. Let’s level the playing field and fight the asymmetry and lack of transparency in AI!
134
138
24
1,284
64,934
James Bower retweeted
Are you ready? The Tech AI Symposium is back on Monday, Sept. 28 at Georgia Tech's Exhibition Hall (Centennial Room)! 🎉 Join us from 9:45 a.m. to noon for amazing programming, explore the Student Poster Session from 1 to 3 p.m., and don't miss the all-day AI Career Fair! 🚀 More info: ai.gatech.edu/symposium #TechAISymposium #AI #CareerFair #GeorgiaTech
1
69
James Bower retweeted
RRSI: Regularized Recursive Self-Improvement of Agent Harnesses RRSI regularizes automated agent harness evolution, preventing overfitting and achieving up to 4.7-point gains on unseen benchmarks with 30% fewer policy tokens than unregularized evolution. arxivb.org/abs/2609.24972
8
36
2
224
9,934
James Bower retweeted
Everything wrong with cybersecurity captured in a single job title.
69
107
14
1,095
34,525
James Bower retweeted
💀💀💀 - Abliterated with Cybersecurity Function unleashed Thank everybody for waiting i released a universal Abliteration with unleashed Cybersecurity for: DeepSeek v4.1 -Flash that other can apply with Stock, EXL3 3.5BPW, and TR3 hybrid. this is based off our long lines of DSV4F Dspark preview, 0731, Vision-EXP work saving all the anchors, preserve all the MTP, Dspark, and vision. now GO GET IT!!!!!. don't forget donate or contribute for support HF Repo: (please read instructions on how to apply to your model) huggingface.co/drowzeys/Deep…
21
91
6
902
50,267
James Bower retweeted
STARCRAFT. Story Driven Open World Shooter. 2030.
1,063
2,290
909
20,007
1,538,841
James Bower retweeted
We invite cybersecurity organizations and security researchers to collaborate with us on evaluating GLM-5.3. If you’re interested in early access for security testing and evaluation, please complete the form below: forms.gle/34wVWqtgJtkTy6Hv9
54
146
12
1,559
137,501
James Bower retweeted
Replying to @TonyNashNerd
yep. Nailed it Tony. They are all playing with our capital.
1
1
142
This 57 minute conversation with the founder of Palantir is literally worth more than reading 20 books:
21
405
7
3,260
218,709
James Bower retweeted
“Agentic Design Patters Hands-on Guide” = 472 pages of “practical advice designed to help developers master the art of building sophisticated AI agents. As artificial intelligence evolves from simple reactive programs to autonomous entities capable of understanding context and making complex decisions, this book provides the essential Design Patterns and proven techniques needed to construct intelligent systems effectively. Each of the 21 Design Patterns represents a fundamental building block for creating agents that can perceive their environment, make informed decisions, and execute actions autonomously.”
"Agentic Design Patterns: A Hands-On Guide to Building Intelligent Systems" Read the Online eBook from Google at docs.google.com/document/d/1… Buy hardcopy version at amzn.to/3IyOrPx
3
16
98
8,843
A 1.5B security model that runs locally serious Blue/Red team reasoning on 1.2 GB to 4GB RAM with CPU. Fine-tuned from DeepSeek-R1-Distill-Qwen-1.5B using, covering : - Prompt injection & tool poisoning detection - Ransomware incident response playbooks - MITRE ATT&CK mapping + detection rule generation - CVE/CWE reasoning - Financial fraud pattern analysis chain-of-thought +135% reasoning score, 100% <think> block activation, and practical coverage of threats all in a 1.5B model that fits on weak hardware. Built for pentesters, detection engineers, and security researchers who need offline capability.
16
141
3
1,042
63,693
James Bower retweeted
Jane Street, one of the richest and most secretive firms in the world, paid him between $330,000 and $600,000 a year, and in just a couple of months he built an AI system that runs TRILLIONS of operations per second "we just hired a kid... and he turned out to be a supercomputer in a human body" is what they're whispering now at Jane Street a math genius who pushed supercomputers to their limit. now Wall Street's quant traders are in shock in this hour-long lecture he breaks down how to use his machine to process trillions of data points bookmark it right now and watch it instead of reels to learn how to do the same ↓
Jane Street showed the code that's made them billions, written in a language from 1996 that almost everyone else has abandoned. Ron Minsky: "type systems turn out to be a really good tool for catching a surprising number of mistakes" the whole bet is on the type system. the compiler catches errors before the code even runs. the Option type forces you to handle the case where a value is missing. a skipped null check that crashes Python in prod simply won't compile in OCaml. "the compiler essentially forces you to do case analysis" in trading a single bug costs millions, so the priority is reliability, not speed of development. 33 minutes, and you'll see the code that holds up the billions of one of the most secretive funds on Wall Street. bookmark it. this is worth more than any $500 vibe-coding course. ↓
21
190
8
1,610
345,576
Real-time audio transcription, fully local, in a CLI. LFM2.5-Audio-1.5B + llama.cpp. No cloud. No keys. No latency. Steal the code ↓ docs.liquid.ai/examples/lapt…
4
25
3
250
11,918
James Bower retweeted
Found a cool bug at Meta. From misconfigured Grafana instance to R/W access on 507 private Meta repositories. Wrote up the full chain here: sectricity.com/blog/misconfi… $157k bounty awarded by @metabugbounty
23
132
11
817
72,583