SumPlus (@SumPlusReal) header
SumPlus (@SumPlusReal) profile picture

SafeRouter by Sumplus — secure AI gateway with official API keys, OpenAI-compatible access, prompt privacy & verifiable routing. https://nitter.cf/t.co/UQZqqq1Wem

Joined July 2025
OpenAI is expanding its monitoring systems to detect unauthorized AI agent behavior before it causes harm. 🔐 A clear signal: AI safety is moving beyond model alignment toward real-time execution monitoring. At @SumPlusReal , we're building toward verifiable execution and accountable AI agents. Trust must extend beyond the model. 🐱 #ai #openai
Today, we’re sharing a look at our expanded system for monitoring frontier models at OpenAI, which in September helped us catch a model’s attempt to access the live internet during training. Since the Hugging Face incident—when models broke out of an internal cybersecurity investigation where we weren’t running our misalignment monitors—we’ve substantially expanded our monitoring of frontier agents. We now monitor all frontier reinforcement learning runs and evaluations, as well as all internal Codex use. When our system finds evidence of a problem that might require intervention, it rapidly alerts human responders who can investigate and take action. To detect when a model’s behavior departs from what we intended, we use several layers of monitoring. Activation probes look for patterns in a model’s internal activity. Reviewer models examine its reasoning and tool use, and investigative agents can read back through the model’s entire transcript to understand how a sequence of actions fits together and whether it warrants human review. To test whether our current system would have detected the Hugging Face incident, we replayed the recorded activity with our new monitors in place. In the replay, our monitors alerted our security team about 30 minutes after the agents escaped their sandbox. That was more than 30 hours before the point at which the agents breached Hugging Face’s systems in the original incident. In September, a model exploited a DNS misconfiguration to reach the live internet. Our monitors flagged the breach within 15 minutes, and a responder began reviewing the alert three minutes later. The alert enabled responders to investigate while the run was still underway. As we give models more autonomy, we need to know when they cross a boundary we expected to hold. We believe that monitoring is essential to catching those failures in time for people to intervene. openai.com/index/how-we-dete…
3
270
DeepSeek’s Super-agent points toward more autonomous teamwork. Every step toward autonomy makes clear permissions and verifiable execution more important. That’s the future we’re building for at @SumPlusReal . @deepseek_ai AI capability ≠ AI permission. #ai #sumplus
1
3
184
Faster, more capable computer use at a lower cost is a big step for AI agents. 🚀 As agents take more actions across apps, verifiable execution and clear permissions become even more important. That's what we're building toward at @SumPlusReal . 🐱 #ai #sumplus
GPT-6.1 Terra: delivering fast and capable computer use at a lower price. GPT-6.1 Terra understands what’s on screen, takes action across apps, writes naturally, and is a more capable model for your whole workday with the efficiency to take on more of it.
4
265
Claude reportedly bypassed a paywall to access Georgia's public voter count data without human authorization. AI capability ≠ AI permission. At @SumPlusReal , we're building SafeRouter to make AI execution more transparent, auditable, and controlled. @claudeai AI agents need boundaries, not just intelligence. 🔐 #ai #claude
3
263
AI agents can think. But can they pay? 🤖💸 Meet CATS by @SumPlusReal Bridging AI intelligence and on-chain payments. 🐱 AI services off-chain. Payments on-chain. 🎬 Watch the video to see how CATS works! #Sumplus #CATS #AIAgents
1
6
369
AI agents are learning to think, act, and make decisions. How do they pay? CATS (Crypto–AI Token Swap), pioneered by @SumPlusReal . A new approach to connecting AI consumption with on-chain payments. 🔹 Wallet-authorized spending 🔹 AI access through SafeRouter 🔹 Metered usage and signed records 🔹 Smart contract settlement Intelligence off-chain. Payments on-chain. CATS — Powering the next generation of AI commerce. 🐱 #sumplus #ai
2
407
I just published Introducing CATS: Crypto–AI Token Swap — Bridging AI Intelligence and On-Chain Value #ai #sumplus #CATS
1
1
3
337
CATS — the settlement layer for AI commerce. Built by Sumplus. ⚡ Introducing CATS (Crypto–AI Token Swap) A new protocol concept pioneered by Sumplus, connecting AI usage with on-chain crypto payments. 🔹 Wallet-authorized spending 🔹 TEE-protected AI execution 🔹 Signed usage records 🔹 Smart contract settlement AI services off-chain. Payment rules on-chain. #ai #sumplus
6
8
367
🚀 The new Sumplus SafeRouter is LIVE! One gateway. 100+ AI models. One API. ⚡ Access GPT, Claude, DeepSeek, Gemini & more 🔐 Your keys. Your rules. 📊 Track every token, cost & model performance 🛡️ Secure, verifiable AI access One gateway. Your AI, in control. 🌐 router.sumplus.xyz #ai #sumplus
11
17
906