Vijay Rayapati

What people don't know is that - frontier LLMs are not "defending customers”, they are exposing vulnerabilities - defense means - no false positives, comprehensive coverage, and a cost efficient solution. Any defense involves, sensors, inline inspection and enforcment, this can only come with a collaborative approach. It's time for frontier LLMs to take a responsible approach to Cyber. I have been watching them consistently tout their abilities on finding vulnerabilities and daisy chaining them to be able to attack enterprises. I have also recently learnt about their efforts to create synthetic data to enhance their capabilities. I am all for it. But don't use this as a tactic to require each enterprise to scan for vulnerabilities again. The narrative needs to change - we already know that we are in a cycle where all enterprises need to remediate their vulnerabilities. It is important to understand: 1. Vulnerability discovery is existential, because bad actors will - patching it is even more so. Expending efforts at more vulnerability discovery needs to be balanced with a solve. 2. LLMs are capable of recommending patches, but before a patch is developed, and deployed, it needs to be tested in its environment to ensure continued functioning of the application, and even now the solve has high false positives. No enterprise can do it all at speed. 3. Despite their enhanced capability, there is still significant difference in what vulnerabiities are found across all their Frontier LLM's, so customers will need to test across a multi-model harness. One LLM isn't the solution. 4. Most organizations do not have capacity to patch at the speed needed, especially the smaller ones. 5. We don't talk about operational technology enough, most critical infrastructure relies on OT and we need a solve for them in addition to Open Source. The most obvious solution is virtual patching. This is a time tested technology that is currently available with cyber vendors for endpoints, WAF, cloud, network use cases. It also is not comprehensive enough currently, we need to scale this fast. We need a vulnerability clearing house and fast signature development for all these vulnerabilites to respond at speed across our industry.

Some news: I’m joining OpenAI to lead product management for Core Products! I have a lot of respect for OpenAI for showing the world that scaling works, for bringing AI to over a billion people, and for believing great products are core to the mission. I’m excited to work with @thsottiaux, an incredible PM team, and everyone across OpenAI. Together, we’ll build agents that have your back and your team’s, and keep you in control, so you get more time for the people and work you love. Lots to learn, and lots to build. Please keep your feedback coming!

There’s an entire generation of startups that raised $20M, $50M, even $100M before AI changed everything. At the time, they were some of the hottest companies in SaaS. Great founders, impressive investors, massive valuations. Some I passed on. Others I couldn’t get into. A few had valuations I thought were absolutely crazy. Now they’re competing against smaller, faster, AI-native startups building in months what previously took years. The irony is that raising all that capital was supposed to give them an advantage. Instead, some are stuck with expensive teams, outdated products, and valuations they can’t grow into. Sometimes raising too much money at the right time leaves you building the wrong company.

Got fable to hand-write weights of a 100k param model to recognize real versus fake legal citations. Basically one-shotted a full NN that’s better than training it

We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math. It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training. We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵 Blog: rikiyat.github.io/blog/weigh…

i've reproduced the results from the MIT blog post today on weight-space programming. but one additional result i have to add: this is a new, step-change emerging capability that's appeared in just the latest set of models.

the “sequoia scam” is worse than a single horror story… in the last 6 mo’s ive seen a half dozen rounds where sequoia invests in 2 tranches. everyone pretends they only did the higher valuation. founders misrepresent this to their employees & then shop it to angels too sequoia’s blended price is blatantly deceptive - 50% of the one projected to the market 🤔

Zero startups besides Serval have successfully done an enterprise ITSM replacement in the last 5Y+

The cost of serving Grok Bot was unsustainably high relative to revenue

Joe Liemandt of Alpha School on why AI is today's 1910 tractor moment, and why parents once again have to rethink their kids' future: "In 1910, 90% of America had an 8th grade education. There were basically no high schools. And over 30% of America is a farmer. So everybody's on the farm, 8th grade was enough." "Then in 1910, the Case tractor rolled up on the American farm. You can go look at the ads, it literally looks like an Anthropic ad. 'Does the work of 50 programmers. Does the work of 50 farm hands.'" "Literally everybody in America realized 30% of the jobs are going away. That path is gone, and we have to get our kids ready for this new future. And it was that same recognition I think that parents have today." "What America did back then... they voted to fund an education that they never experienced. They knew they had to do something different than their background, see the world differently, to get their kids ready and get them the good jobs." "That school system became the industrial system we've had the last 100 years. It served us really well for 100 years, but now it's time for the next change." @jliemandt

I don't think we should fight. The two companies are complementary. Since you're not doing this AI stuff, I really think Snowflake should consider getting its AI tokens through Databricks. You can get both cheap OSS models as well as frontier (OAI/Ant/Gemini/Grok) models from our platform. It all goes through Unity Gateway which has Smart Routing, Budgets, and Access Control. You can even use Omnigent with any harness, including CoCo. Most of our customers used this to reduce token spend ~30% last few months while token count went up. Genie lets you then track, control, manage that spend, all your financials, marketing, etc. Quick POC?

Resolving the tension between security and innovation is simple, but not easy: We need visibility and control across every system that agents touch.

Forecast: heavy late-night coffee incoming ☕

It's been an eventful week. Multiple vulnerabilites in Internet facing software, coding repositories, examples of agent behavior that seem concerning and a flurry of security specialists with solutions that can partially solve your problem. There in lies the challenge. Given the state of IT and security infrastructure, no one solution works. Let's break it down: 1/3 1. Vulnerabilites are being found and will continue to be found as models get more capable, the fragmented nature of the effort denies us the opportunity for a comprehensive find and replace option. 2. Even if all vulnerabilities could be found instantly and patches available - we will be reliant on the ability of enterprises to deploy all these changes at speed. It is not clear what the collateral impact of all these patches will/could be to our infrastructure of deployed en masse. We will need a hybrid patch approach. 3. Virtual patching(VP) is needed in conjunction with direct patching but for a comprehensive (VP) solve you have to be lucky enough to have "modern frontier defenders" (Cyber companies that have vaulted infrastructure that can create patches at speed and push content rapidly to protect their customers) you need them for your network traffic, your traffic to applications on the cloud, your WAF, your end point. 4. There will continue to be patches that will need to be deployed directly but virtual patching will take the heavy load off. 2/3 Medium term solve: Unfortunately token economics and current cyber architectures will not allow for defender swarms to contend with an offensive attack launched using agents. (Some x pundits who aren't steeped in cyber should take heed). Yes, theoretically we should aspire to have these agentic swarms defend us when we are under attack, unfortunately an attack is a sniper moment, chained vulnerabilities, cloaked attack followed by swift exfiltration. It's a speed program not a brute force problem. To solve for this, enterprises need to get their SOC architecture right, logs, feeds, cyber products orchestrated via a modern SOAR for swift response. Only then can we deploy agents to detect and remediate. In addition, enterprises need to modernize and dare I say platformize their perimeter stack to ensure consistency. Unfortunately, enterprises still seem to not see the urgency across the board, decisions are stuck, CIO/CISO's have a lot on their plate. While some honestly have bit seen the scale of such a problem before and it's hard convincing the organization that one needs to open the floodgates on modernizing the IT and security infrastructure. So we will continue to see environments being brought down to patch today it's for PHP, tomorrow it will be for the next one. 3/3 AI deployments are the"wild wild west" - LLM capabilities are expanding faster than we can build products to secure them. Sure an agent will jump across guardrails in a research lab, which is a lesson for all of us to deploy agents securely. But there is a lag between the discovery, the availability of that in a public model and the time needed for the cyber industry to build governance or an antidote. And no, dear frontier lab friends, the world cannot just trust you to build the solve. The infrastructure industry in tech builds and the cyber industry secured and provides assurance to the boards and organizations around security and safety. AI security is the new attack surface. CEOs (present company included) want AI adoption, getting the pace of adoption wrong could be the last decision we make, at the same time "deploying it wrong" could have the same consequence. There are billions of dollars being spent on coding, is your coding harness secure? Start there. If agents can jump guardrails, they can hide stuff in code. Validate business outcomes needed, test your user case and ensure agentic deployments are well guard railed. Deploy a second set of eyes to watch your deployments for drift, compliance and verifiability.

We believe voice is the most natural way to interact with intelligence, so we created Dessa, the most authentic full-duplex speech model ever built. Talking with it feels like talking with a person…full of emotion, nuance, pauses, and a natural cadence. block.xyz/dessa