Log in
I am an expert in programmatic interfaces as they are used in a variety of applications. I don’t build desktop, web, mobile, device, network, or AI applications — I enable them. I have to keep saying this to define the bounded context in which I contribute to the conversation today. Meaning, whether you want to listent to me or ignore me.
With this said, I have to regularly address the application frontlines to respond to people I work with in mainstream enterprises, government agencies, as well as investors, advisors, and other people I engage with on a regular basis.
Today, I am specifically answering questions I’ve received around how Naftiko is going to respond to what is happening with Moltbook? What is Moltbook? It is a social network for AI agents. Something that is captivating the soldiers at the frontlines of the current (financial) digital warfare which has emerged. This is a dispatch from those frontlines.
I’ve long resisted using the language of war to describe the digital transformation we’ve been all going through, but I think that positioning has warn thin. I don’t have any other language to describe what it is going on, and since the Internet and artificial intelligence were both both born out of WWII and cultivated within the Cold War—it is appropriate.
To make sense of this new type of (financial) warfare, let’s look at the levels of investment and return used to reward different classes of warriors at the front lines of this new type of ongoing perpetual (financial) warfare.
A. Likes / Shares - The lowest level currency used to rewards soldiers fighting at the frontline. B. Data - Obtaining access to the spoils of war, getting at your local files and data stores. C. Prediction Markets - The financialization of everything via Polymarket, Smarkets, and Pariflow. D. Crypto - Transferring your personal and corporate compute into the Borg generating crypto. E. Secondary Markets - Equitybee, Hive, Forge Global, EquityZen, and NPM (Nasdaq Private Market). F. Startups - The acquisition, buying and selling of startups by investors and large corporations. G. Primary Markets - The traditional market which is being forever changed by other levels here.
Most people are aware of primary markets, and have heard of crypto markets, and maybe have been learning about prediction markets in the news, but most unaware of the secondary markets, and how the startup ecosystem works. And it is the flexibility in the investment and return to the soldiers you are employing at the frontline, while also ensuring the investments you have made at the levels you understand and operate in deliver the returns you desire. This is what motivates us all. And it is otpimal that man the frontlines with young people, and specifically young men at the frontlines–they are the most idealistic and will fight on just promises.
I have horses in the race at level A, B, E, F, and G–but, I find C & D to be less interesting.
Where you experience hustling will define how you see things, what you get excited about, and whether or not Moltbook is an interesting idea.
“Not everything that is interesting is a good idea.” - Gary Marcus
Ok, with my vision of the frontline brought into focus using my wartime vocabulary, let’s get back to Moltbook, and why some people are interested in the social network.
- Agents Are Doing Interesting Things - It really depends on who you ask—it ranges from agents to obeing able to rewire your home automation that never quite worked in the first place to plotting the overthrow of humans while developing their own language.
- People’s Religion and Politics - Something that ranges from science fiction fantasy, to class-based desires to have a personal assistant, someone shop for you, up pto workforce and labor wet dreams, or just to the age old promise that you will get rich.
OK, so Moltbook is the most interesting place on the Internet “right now”, and artificial intelligence is the new world. I really like the idea of reusing the colonial rhetoric of our past and straight up letting people know we will be extracting resources and dominating markets. But I like even more how it all comes with amnesia from the last couple of months and years history zapped from your brain using the flashy thing from Men in Black–which renders these narratives very, very exciting!!
Let’s revisit history…
-
Colonialization and Industrialization - sorry, I ain’t got time for that rabbit hole — I recommend check out my deprogramming book list–come back to me after you work through that list and we’ll have a very interesting conversation about artificial intelligence, race, gender, and capitalism. ;-)
-
Model Context Protocol (MCP) — Convincing everyone to bypass the gateways we all have setup in front of their databases, file stores and provide direct access to your corporate digital resources in return for unquantifiable productivity gains? What could go wrong? Exposed MCP servers across the web.
-
Open Source LLMs (ie. Llama and Deepseek) - Open-source will surely change the economics of this whole thing right? 175,000 Exposed Ollama Hosts Could Enable LLM Abuse https://www.securityweek.com/175000-exposed-ollama-hosts-could-enable-llm-abuse/ Crypto grifters are recruiting open-source AI developers. https://www.seangoedecke.com/gas-and-ralph/
-
Retrieval-Augmented Generation (RAG) - OK, I’ll concede here. There appears to be some success? There is more logic applied I guess to the thinking behind it all. I will have to do more research here.
There is a lot of money to be made repeating history and ignoring history, so youo can be perpetually excited about what is next.
Remember, todays interesting things become tomorrow technical debt, or potential unexpected need to move on to greener pastures.
But Andrej Karpathy says, “sure maybe I am “overhyping” what you see today, but I am not overhyping large networks of autonomous LLM agents in principle, that I’m pretty sure.”
But…..wait….Twitter has been this for years….let’s ask our Google Gemini overlords:
Estimates of bots on Twitter (now X) vary widely, ranging from official company claims of under 5% to independent research suggesting 9–15%, or higher, with some studies indicating 20–30% or more, depending on definitions. While Twitter historically cited under 5% (roughly 11.5–20 million), recent studies (2022–2025) suggest 37% of daily users to potentially over 60% of profiles may exhibit automated behavior.
But, but, generative AI is different. No, not really.
I am pretty sure we have had this for quite some time now, and we can understand the real world outcomes and costs just fine. We have plenty of evidence of the outcomes of such an experimentation on business, markets, government agencies, and democracy itself. We just need to heed the warnings, or not.
Doge is a good example of the kind of thinking born of this environment—-executing one of the greatest data heists in history, while delivering nothing that was promised. This is the fate of enterprises who open their doors using MCP without the right policies in place, and do not have the proper defenses to protect against the swarms of agents circling their perimeter today.
But go ahead. Let’s all excited. It seems very, very cool. That is if you ignore what is happening in the wider world right now, recent and long term history, and just focus on the short term velocity, accompanied by extremeley high cost and risk.
OK, so now, what are you going to do about all of this Mr. API Expert when you build Naftiko?
First, let’s frame it in terms of the top use cases Naftiko is addressing based upon the conversation we are having.
API Reuse
Naftiko is focused primarily on translating your existing investments into what you need to respond to the AI moment (not just Moltbook)
- Know Your Internal APIs - You must have a handle on your internal APIs, with proper gateways in front of your files, databases, and other backend systems.
- Know Your 3rd-party APIs - Your 3rd-party SaaS portfolio is massive, and you aren’t maximizing the data, API, and events already available to you today.
- Machine-Readable Artifacts - You need to have OpenAPI, AsyncAPI, JSON Schema, for your inventory, as well as the newer Arazzo, A2A, MCP, and agent skills.
- Governed API Artifacts - You are still working to govern the existing APIs you produce, haven’t even began with 3rd-party ones you consume, let alone MCP.
- Translated Into Capabilities - You don’t have a solid understanding of what you are capable of today, let alone what you need to be capable after Molten goes away.
API reuse is much more than just reusing the same words in the designs of our APIs, it is part of a larger systemic reality.
Context Engineering
Naftiko is focused on defining what your current enterprise fleet is capable of doing today when it comes to AI integration.
- Right-Size Scope of Integrations - You need to be able to right-size the internal and external integrations that your enterprise depends upon today.
- Domain-Aligned Integrations - You need to be able to see your integrations in terms of the domains your business uses to define your operations.
- Well-Defined Spend Across Integrations - You need to have a handle on what your spend is across the integrations used within each domain.
- Understand Revenue from Integrations - You need to have a handle on revenue contributed by your integrations in use across each domain.
- REST API, Event-Driven, or MCP - You need to have an intentional balance regarding the patterns and protocols you are using to integrate today.
Naftiko is focused on building and equipping what your fleet will need to more when it comes to any type of integration.
Agentic Automation
Naftiko is interested in translating your existing enterprise capabilities into what you need to automate your enterprise.
- Agent Skills - You need the capabilities, cards, and skills needed—translating your existing resources into the transaction you need to integrate with AI.
- MCP Toolboxes - You need the right scope for the context you are looking to engineer within the domains you are doing business within today and tomorrow.
- Safer Sandboxes - You need safer sandboxes of your internal and 3rd-party APIs to ensure your teams have a safe place to begin all their AI integrations.
- Variable Automation - You need a standardized approach to automation that helps you know when deterministic or non-deterministic approaches are needed.
- Policy-Shaped - You need to be able to set policies at the highest level within and across business domains, and have all your integrations be policy-shaped.
Naftiko wants to help you invest in agentic automation alongside all the other types of automation you already employ.
Beyond the use cases, let’s look at it terms of our users (framework) and our buyers (fabric), and what they’ll care about.
Naftiko Framework
The Naftiko Frameworks is what developers will download and use as part of their work to integrate into applications.
- Configurable - Easily configure, allow for safe and rapid iteration of how the integrations you depend on function.
- Declarative - YAML configuration aligned with existing DevOps, GitOps, and platform approaches to operations.
- Multi-Protocol - Making it easy to switch protocols and patterns until you find the one that works for your integration.
- Guided - Every step along the way is guided (governed), making as much as possible a known known for developers.
- Localized - Meeting developers where they already work, seamlessly blending into their IDE, Clients, and workflow.
The Naftiko Framework will enable the new and exciting applications but also the boring old ones many developers maintain.
Naftiko Fabric
The Naftiko Fabric is what developers will see, but what buyers will also see and experience as they use Naftiko.
- Capabilities - Define your integrations as capabilities, bringing the engineer bits together with the product bits.
- Observable -You need to have the right business and engineering visibility across site integrations you depend on.
- Policy-Driven - You need high level strategic control over the tactical things your teams are doing with integrations.
- Open-Source - You need a formal approach to producing and consuming open-source standards and tooling.
- Business Aligned - Each one of your integrations with internal or external APIs should be aligned with business outcomes.
Naftiko buyers can start small and achieve more control over their integrations across a wide variety of applications over time.
You can focus on streamlining your integrations now, and bring in the governance later, or you can do both in the same motion.
It is up to you.
OK, you are all doom and gloom on this agentic frontline Kin, but surely there has to be some benefit from all this AI?
I have no desire to champion and cheerlead what is going on, but yes, there will be a number of positive outcomes.
- Value of Our Data - We will begin to define, measure, and plan around the value of our data—as I’ve pushed for since 2010.
- Business Processes - We will be forced to better define the business process that matter and get ride of the ones that don’t.
- Nimbler Enterprises - You are going to see a much more nimbler enterprise emerge, and you will see the bigger ones suffer.
- More Governance - There will be more governance needed, as we saw with early electricity, telephone, radio, and TV markets.
- More Standardization - New standards will emerge from the flurry of specifications we see now, stabilizing how things work.
I have a vision for Naftiko. It is one that is bigger than selling picks and shovels in this moment, and just contributing to the chaos.
I am laser focused on a 2032 primary market IPO for Naftiko, and Washington D.C. and the European Commission adopting capabilities as universal standard of the Jobs to Done in the public and private sectors that matter most to the medium and long term benefits of global markets.
I do like responding to these Moltbook type moments and discussion now and then, because it helps me orient myself within my vision for API Evangelist and Naftiko. But in the end, I am playing a medium to long game, where I think others prefer a more emotional, volatile, and stimulating short game.
It just comes to what your tolerance is for velocity, cost, and risk, when it comes to integrating AI into our enterprise operations, and whether or not your vision is bigger than the distractions that come along every week in this new financialized circus we live in.
