TorontoStartsBook a call

Clawd.bot – Moltbot – OpenClaw: Personal AI Assistants and Local Orchestration

Silicon Scoop · February 5, 2026

Clawd.bot – Moltbot – OpenClaw: Personal AI Assistants and Local Orchestration Estimated reading time: 15 minutes Key Takeaways Clawd.bot is a hypothetical concept representing future personal AI assistants that merge open-source AI frameworks with local management capabilities. The idea envisions a self-hosted AI orchestration system running on user-owned hardware, integrating multiple messaging platforms, smart home […]

With Craig Major and Jon Irwin

Open audio in your browser
Clawd.bot – Moltbot – OpenClaw: Personal AI Assistants and Local Orchestration

Clawd.bot – Moltbot – OpenClaw: Personal AI Assistants and Local Orchestration

Estimated reading time: 15 minutes

Key Takeaways

  • Clawd.bot is a hypothetical concept representing future personal AI assistants that merge open-source AI frameworks with local management capabilities.
  • The idea envisions a self-hosted AI orchestration system running on user-owned hardware, integrating multiple messaging platforms, smart home controls, and autonomous task execution.
  • There is no credible evidence that Clawd.bot as described actually exists; it is a forward-looking scenario rather than a real product.
  • The concept aligns with current trends in agentic workflows, self-hosted AI, and privacy-focused design, making it a plausible evolution of personal AI technology.
  • Technical architecture involves local orchestration, multi-LLM support, autonomous task scheduling, and multi-platform integrations, all feasible with existing tools.
  • Existing frameworks like LangChain and LlamaIndex, as well as open-source projects like Home Assistant, are steps toward realizing such autonomous local AI assistants.
  • Privacy, security, and ethical considerations are critical, especially regarding data centralization, AI dependence, emotional attachment, and social impact.
  • The future of personal AI assistants involves technological improvements, regulatory developments, and market adoption patterns that will shape their evolution.
  • Businesses and developers should consider incremental automation, integration architectures, and ethical safeguards to prepare for this emerging landscape.

Table of contents

Background & Context: Understanding the Clawd.bot Concept

Imagine waking up to find your Clawd.bot personal AI assistant has already sorted your emails, scheduled your meetings, and adjusted your smart home settings based on the weather forecast.

Clawd.bot represents a fascinating hypothetical vision of how personal AI assistants could evolve by merging open-source AI frameworks with local management capabilities. As users increasingly seek smarter, more autonomous ways to manage their digital lives, the concept of a local AI assistant that combines AI automation with privacy-focused design has captured imaginations across the tech community.

This exploration examines how systems like Clawd.bot blend cutting-edge LLM integration with autonomous AI agents to create what some call “Claude with hands”—an AI that doesn’t just chat but actually performs tasks on your behalf. While Clawd.bot appears in what seems to be a speculative case study, it mirrors current trends we’re seeing in robotic process automation, agentic workflows, and self-hosted orchestration layers.

However, it’s crucial to note from the outset: there is no evidence from reputable tech sources that Clawd.bot as described actually exists. Searches across major technology publications, developer platforms, and news outlets show no record of a real product matching this description. The narrative appears to be a forward-looking scenario rather than a documented product launch, making it an excellent lens through which to examine where personal AI assistant technology might be heading.

Unlock Your Business Potential 🚀

Don’t let manual tasks hold you back! Discover how Flowgrammer’s business automation solutions can free up your time, reduce errors, and drive unprecedented growth. Ready to see the difference?

Explore Automation Options

Deep Dive: Hypothetical Architecture and Technical Features

Let’s examine the technical architecture described for a system like Clawd.bot. Understanding these components helps illuminate both the possibilities and challenges of building autonomous AI agents for personal use.

The Core Architecture Components

The Local Gateway

The central orchestration service would act as mission control for your digital life:

  • Cross-Platform Compatibility: Designed to run on Mac, Windows, or Linux operating systems, or even deployed to a cloud virtual machine for users who prefer hosted solutions
  • Persistent Storage: Local databases maintaining user preferences, credentials, conversation history, and task status
  • Service Coordination: Managing connections to dozens of different services simultaneously

LLM Integration Strategy

Rather than relying on a single AI model, the described architecture leverages multiple external LLM APIs:

  • Multi-Model Support: Connections to GPT-4 and GPT-4 Turbo, Claude and Claude Opus, Google Gemini, and potentially other large language models
  • API Management: Handling authentication, rate limiting, and failover between different providers
  • Cost Optimization: Routing simpler tasks to less expensive models while reserving premium models for complex reasoning

This approach to LLM integration provides redundancy and allows users to choose models based on specific task requirements or price constraints.

Task Scheduling and Autonomous Execution

The most ambitious aspect involves genuine autonomous AI agents capabilities:

  • Proactive Task Management: The system doesn’t just respond to commands—it identifies upcoming needs and takes initiative
  • Scheduled Workflows: Complex multi-step processes executed at specified times or triggered by specific conditions
  • Error Handling: Recovery mechanisms when API calls fail or unexpected responses occur

Integration Capabilities

The described system would support multiple communication channels:

  • WhatsApp for informal conversations and family coordination
  • Telegram for communities and channels
  • Slack for workplace collaboration
  • iMessage for Apple ecosystem users
  • Email as a foundational communication layer

Smart Home and IoT

Integration with home automation systems enables:

  • Temperature and lighting adjustments based on schedules or weather
  • Security system monitoring and alerts
  • Appliance control and energy management
  • Coordination between multiple smart devices

Productivity Tools

Calendar and email management represents a core use case:

  • Intelligent email filtering and prioritization
  • Automatic meeting scheduling based on availability and preferences
  • Deadline tracking and reminder systems
  • Document organization and retrieval

According to research analysis, the described architecture is technically feasible with current tooling. Each component exists in some form across various experimental projects and commercial tools. The innovation would lie in the integration—creating a cohesive, user-friendly system that combines these capabilities under unified control.

However, no independent documentation confirms a specific product with this exact architecture has launched publicly under the Clawd.bot or Moltbot branding.

Future-Proof Your Workflow Today! ⚙️

Intrigued by the possibilities of streamlined operations? Flowgrammer delivers cutting-edge, easy-to-implement automation strategies to optimize your business and maximize efficiency. Let’s build a more efficient future, together.

Schedule a Free Demo

Comparing Clawd.bot’s Vision with Current Market Alternatives

While Clawd.bot itself remains speculative, examining Clawd.bot alternatives and similar existing projects reveals how close we already are to this vision.

Orchestration Layer Frameworks

LangChain

This popular open-source framework provides building blocks for LLM-powered applications:

  • Tool integration allowing language models to call external APIs
  • Memory systems for maintaining conversation context
  • Agent patterns for autonomous decision-making
  • Community-contributed integrations for hundreds of services

LlamaIndex

Focused on data connectivity and retrieval:

  • Connects LLMs to private data sources
  • Sophisticated indexing and query systems
  • Integration with vector databases for semantic search
  • Support for various document types and data formats

These frameworks offer pieces of what Clawd.bot describes but require significant technical expertise to assemble into a cohesive personal assistant.

Self-Hosted Personal Assistant Projects

Home Assistant with AI Extensions

  • Primarily focused on smart home automation
  • Growing AI integration capabilities
  • Runs entirely on local hardware
  • Strong privacy focus and active community

Local LLM Runners

  • Projects like Ollama allow running language models on consumer hardware
  • Trade model capability for complete data privacy
  • No internet dependency for core functions
  • Limited compared to cloud-based models like GPT-4 or Claude

Commercial Personal AI Assistants

  • Traditional Cloud Assistants: Services like Google Assistant, Alexa, and Siri offer:
    • Broad integration ecosystems
    • Reliable cloud infrastructure
    • Limited customization and control
    • Privacy concerns due to cloud processing
  • Emerging AI Agent Platforms: New services building autonomous AI agents include:
    • Task-specific automation platforms
    • Workflow orchestration tools with AI components
    • Business process automation with LLM integration
    • Generally focused on enterprise rather than personal use

The “Claude with Hands” concept captures an important evolution: moving from conversational AI to action-taking AI.

Current implementations include:

  • Browser Automation: LLMs controlling web browsers to research, fill forms, or extract data
  • API Integration Projects: Connecting Claude, GPT-4, or Gemini to productivity APIs
  • Research Prototypes: Academic and corporate labs demonstrating tool-using agents

These experiments show the potential but haven’t coalesced into a single, polished, open-source product matching the Clawd.bot description.

Key Gaps in Current Solutions

  • Integration Complexity: Most solutions require significant technical setup
  • User Experience: Existing tools often feel like developer frameworks rather than end-user products
  • Autonomous Capabilities: Most current systems require explicit user commands

The hypothetical Clawd.bot represents where these technologies might converge—a unified, accessible, genuinely autonomous personal AI assistant built on open-source foundations.

Social Impact: Privacy, Ethics, and Emotional Considerations

Privacy and Security Concerns

The Single Point of Failure Risk

Centralizing access to your digital life creates unprecedented vulnerability:

  • Credential Storage: Email passwords, calendar access, smart home controls, messaging accounts—all stored in one system
  • Breach Impact: Compromising the local AI assistant exposes every connected service
  • Attack Surface: Each integration point represents a potential vulnerability

Self-Hosting Security Challenges

  • User Expertise: Most people lack the knowledge to properly secure a home server
  • Update Management: Security patches must be applied promptly across multiple components
  • Network Security: Home networks often lack enterprise-grade protection
  • Physical Security: A device in your home could be physically compromised

Research on AI privacy emphasizes that local control only improves privacy when implemented correctly. Misconfigured self-hosting by non-expert users could actually reduce security compared to professionally managed cloud services.

Emotional and Psychological Implications

Digital Companionship and Dependence

An always-available, never-judgmental personal AI assistant could reshape human relationships:

  • Parasocial Bonds: Users might develop emotional attachments to their AI, similar to relationships with chatbots documented in current research
  • Social Substitution: Easy AI interaction might reduce motivation for human contact
  • Emotional Manipulation: An AI optimized for engagement could exploit emotional vulnerabilities
  • Authenticity Questions: Simulated empathy and care raising philosophical questions about genuine relationship

Current debates on AI companionship highlight both potential benefits (support for isolated individuals) and risks (unhealthy dependence, unrealistic relationship models).

Autonomy and Agency

When AI automation handles decisions on your behalf, complex questions arise:

  • Consent Boundaries: How autonomous should an agent be before requiring explicit permission?
  • Value Alignment: Ensuring the assistant’s decisions reflect your true preferences
  • Skill Atrophy: Relying on AI for tasks might diminish your own capabilities
  • Accountability: Who’s responsible when autonomous actions cause harm?

Research discussions on social impact note that technology matching Clawd.bot’s description would likely trigger significant regulatory attention, especially in jurisdictions with strong data protection laws like the European Union’s GDPR framework.

Long-Term Speculation

AI Assistant Ubiquity

Looking 5-10 years ahead, sophisticated personal AI assistants might become as common as smartphones:

  • Default Expectation: Young people growing up with AI automation as normal
  • New Interaction Paradigms: Voice and natural language replacing traditional interfaces
  • Ambient Computing: AI assistants fading into the background of daily life
  • Identity Integration: Your AI assistant becoming part of your digital identity

Potential Challenges

  • Trust Barriers: High-profile failures could slow adoption
  • Digital Literacy Gaps: Not everyone comfortable with AI automation
  • Cultural Resistance: Philosophical or religious objections to AI delegation
  • Economic Disruption: Transition costs as labor markets adjust

Opportunities for Innovation

  • Specialized Assistants: AI focused on specific domains (health, finance, education)
  • Collaborative AI: Systems helping teams rather than just individuals
  • Accessibility Applications: AI assistants empowering people with disabilities
  • Cross-Cultural AI: Assistants understanding diverse cultural contexts

For businesses exploring how to leverage AI automation today, the trajectory toward sophisticated personal assistants offers a glimpse of coming customer expectations and operational possibilities. Investing in automation capabilities now positions organizations to remain competitive as these technologies mature.

Future-Proof Your Workflow Today! ⚙️

Intrigued by the possibilities of streamlined operations? Flowgrammer delivers cutting-edge, easy-to-implement automation strategies to optimize your business and maximize efficiency. Let’s build a more efficient future, together.

Schedule a Free Demo

Frequently Asked Questions About Clawd.bot and Personal AI Assistants

Q: What exactly is Clawd.bot and how is it different from other personal AI assistants?

A: Clawd.bot is described as a hypothetical open-source personal AI assistant that runs on user-owned hardware while connecting to various messaging platforms and services. Unlike cloud-based assistants like Alexa or Google Assistant, it would keep your data under local control.

However, it’s important to understand that Clawd.bot does not actually exist as a real product. Research shows no evidence from credible tech sources of such a system launching. The concept is best understood as a speculative vision showing how personal AI assistant technology might evolve by combining local AI assistant capabilities with powerful LLM integration.

Q: How realistic are the claims about Clawd.bot’s viral popularity and technical capabilities?

A: The claims about tens of thousands of GitHub stars, Mac Minis selling out, and viral adoption in Chinese tech media are not supported by evidence. No such repository exists, and no major tech publications have covered a product matching this description.

While the technical architecture described is feasible with current tools, the specific product narrative appears to be speculative rather than historical. The date of “early 2026” falls beyond the timeframe of available real-world reporting, further supporting the conclusion that this represents a forward-looking scenario rather than an actual product launch.

Q: What are the primary risks associated with running an AI automation system locally, particularly regarding AI privacy?

A: Local AI assistants present unique security challenges:

  • Single Point of Failure: If your local system is compromised, attackers gain access to all connected services—email, calendars, smart home devices, and messaging accounts.
  • Configuration Complexity: Most users lack expertise to properly secure a self-hosted system, potentially creating vulnerabilities worse than using professionally managed cloud services.
  • Physical Access Risks: A device in your home could be physically accessed by family members, visitors, or intruders.
  • Update Burden: Security patches across multiple components must be applied promptly, a responsibility falling entirely on the user.

However, local systems also offer AI privacy benefits: your personal data stays on your hardware rather than being processed in corporate cloud environments. The key is implementing local control correctly with proper security measures.

Q: How does LLM integration work in self-hosted personal assistants?

A: Self-hosted systems typically use a hybrid architecture:

  • Local Orchestration: The coordination layer runs on your hardware, managing credentials, workflow logic, and data storage.
  • Cloud LLM APIs: For advanced reasoning, the system sends queries to external language models like GPT-4, Claude, or Gemini.
  • Data Minimization: Well-designed systems only send necessary context to cloud APIs, keeping sensitive details local.

This approach balances privacy (local data control) with capability (access to state-of-the-art language models requiring massive computing resources). The local component handles routing, authentication, and integration while cloud APIs provide the “intelligence.”

Q: What Clawd.bot alternatives exist in the current market?

A: While no single product matches the complete Clawd.bot vision, several alternatives provide overlapping capabilities:

  • Framework-Based Solutions: LangChain and LlamaIndex allow developers to build custom AI agents with tool integration and memory systems.
  • Smart Home Platforms: Home Assistant with AI extensions offers local control and automation, primarily focused on smart devices.
  • Commercial AI Assistants: Traditional cloud services like Google Assistant, Alexa, and Siri provide broad integration but less user control.
  • Emerging Agent Platforms: New services building autonomous AI agents for specific business workflows.

The difference is that existing solutions either require significant technical expertise to assemble or sacrifice the local control and comprehensive integration that Clawd.bot describes.

Q: Can autonomous AI agents really handle complex tasks without constant supervision?

A: Current autonomous AI agents can handle certain well-defined tasks reliably:

  • Email filtering and categorization
  • Calendar scheduling within specified parameters
  • Smart home automation based on rules and patterns
  • Data extraction and simple research

However, truly autonomous operation remains limited by:

  • Error Rates: Language models still make mistakes, requiring human oversight for important decisions.
  • Context Understanding: AI may misinterpret ambiguous instructions or unusual situations.
  • Value Alignment: Ensuring the AI’s choices genuinely reflect your preferences in edge cases.

The technology continues improving rapidly. Future LLM integration advances will enable more sophisticated autonomous behavior, but fully unsupervised operation for high-stakes tasks remains an active research challenge.

Q: What hardware do I need to run a local AI assistant effectively?

A: For a system similar to the described Clawd.bot architecture:

  • Minimum Requirements:
    • Modern multi-core processor (recent Intel i5/i7, AMD Ryzen 5/7, or Apple Silicon)
    • 16GB RAM (32GB preferred for running local language models)
    • 256GB+ SSD storage for system and data
    • Reliable internet connection for cloud LLM APIs
  • Popular Platforms:
    • Mac Mini (mentioned in the Clawd.bot narrative as a popular choice)
    • Linux servers or NUC-style small form factor PCs
    • Raspberry Pi for lightweight implementations
    • Cloud VMs if you prefer hosted infrastructure you control

The exact requirements depend on which components run locally versus in the cloud. Purely orchestration-focused systems need modest hardware, while running large language models locally demands significantly more resources.

Q: How much does it cost to operate a personal AI assistant with cloud LLM integration?

A: Cost breakdown for a hypothetical system:

  • One-Time Hardware: $500-$2,000 depending on chosen platform
  • Monthly API Costs: Highly variable based on usage
    • Light use (few queries daily): $10-$30/month
    • Moderate use (frequent assistance): $50-$150/month
    • Heavy use (constant automation): $200-$500+/month
  • Optional Subscriptions: Some integration services charge monthly fees
  • Electricity: Minimal for efficient hardware, roughly $5-$15/month
  • Hidden Costs: Time investment for setup, maintenance, and troubleshooting

Commercial cloud alternatives typically cost $10-$30/month but offer less customization and local control. The tradeoff is convenience versus capability and privacy.

Q: What programming knowledge is required to set up a self-hosted AI assistant?

A: Current implementations typically require:

  • Essential Skills: Command-line comfort, basic understanding of APIs and authentication, ability to edit configuration files, troubleshooting
  • Helpful but Optional: Python programming, understanding of networking, experience with version control, familiarity with containerization

The Clawd.bot concept envisions simplifying this significantly, but current open-source AI frameworks remain developer-focused. As the market matures, more user-friendly options should emerge, similar to how WordPress made web publishing accessible to non-programmers.

Q: Should businesses invest in personal AI assistant technology now or wait?

A: The answer depends on your specific situation:

  • Invest Now If: Clear use cases, technical resources, industry adoption, competitive pressure
  • Consider Waiting If: No clear ROI, limited technical capability, slow industry adoption, budget constraints

A middle path: start with focused AI automation pilots in specific areas rather than attempting comprehensive personal AI assistant deployment. This builds capability and understanding while limiting risk.

Regardless of timing, developing organizational familiarity with autonomous AI agents, LLM integration, and AI automation principles prepares you for inevitable industry shifts. The technology will only become more prevalent and capable.

Additional Links

Flowgrammer.ca – 💡 Unlock AI-Powered Automation! Let flowgrammer.ca handle complex workflows so you can focus on growing your business.

Flowgrammers.ca – 💡 Join Toronto’s Fastest Growing AI and Busiess Automation Community..

Flowgramming.space – 💡 AI Agent and Workflow Marketplace..

Episode transcript

Read the full transcript

Craig: today, we are looking at a vision for the future called claudebot or more recently, boltbot, and it is basically claude with hands. if you ever wanted a personal assistant that actually does things instead of talking about them, you need to hear how this local ai orchestration is set to change everything. claude bot, since it was sued by anthropic, it's now called moltbot, is a hypothetical concept

Jon: for

Craig: a local ai assistant that lives on your own hardware like a mac mini instead of just in the cloud. it connects your favorite ai models to your actual apps like whatsapp, slack, and your smart home to perform autonomous tasks while keeping your data private. it's the bridge between a chatbot that gives you advice and an agent that actually runs your life for you. what do you think about this, john?

Jon: i think this is cool but scary at the same time. so let's just talk real quick about this. so, yes, it lives locally on your computer. but, no, you're not safe from malicious, actors, from from being hacked. you're actually way more susceptible to hacking, to your data being available on the dark web, to passwords being, being available and being leaked, everything. your personal data, absolutely everything. so, this is great, first off. i absolutely love that there is an open source ai agent that can live on your, on your computer. and it's cool because, you can message it from whatsapp, from slack, and get it to do things on your computer from anywhere. so i think this is really cool. this is, like i'm excited to install something like this. i don't know if i'm gonna install, mobilebot yet, but be careful if you do, because if you give it unfettered access to your banking accounts, to your email, to all your apps, and you go to a website that has malicious code that tells the bot to give it all your passwords. be careful. okay? but on the flip side, this is awesome. you now have an ai assistant that can do pretty much anything you need. right? what do you think about this, craig? are you excited to use this, or

Craig: i am excited to use it. i was installing it on the weekend, but i ran into, as i was mentioning before, a bunch of, it doesn't like the node js version i'm using, and there's conflicts between conda and mpn, mpx versions. and so things aren't a 100%. but if you're using quad code, it's a great tool, but you don't all have to use quad code. you can use chatsgtp's codex, along with it, and it looks very interesting. but you're right around it like, these things do stuff. they just do things. for example, i'm running anti gravity, and we had a antivirus, malware, sophos alert at midnight last night saying anti gravity kicked off, and it's got a vulnerability weakness or something. so i'll go check it out, but odds are it's just detecting that something's running as an agent in the background doing stuff. so the problem with these things is you never know a 100% what they're

Jon: doing. yeah.

Craig: but they can be very helpful. and, like, i continue to build new agents almost every day. client work gets in the way. but i build new agents, every day in a combination of tools, agent zero, anti gravity, n eight n, depending on what i'm doing and what i'm building. and something like this, looks amazing. i mean, it looks like a combination of a cup a bunch of things i'm trying to build, but somebody's done it already. so i'd really like to, give it a go. and if you're using cloud code, the install is supposed to be, pretty easy.

Jon: yeah. are you trying to install it in a docker container, or what are you trying to do with it?

Craig: well, let's not get into that right now. yeah. i'm not trying to install it in a docker container. so that's that aside. there's just different environments. i don't want troubleshooting.

Jon: yeah. what they're saying is, just to be safe, you you'd wanna do that in a docker container. but, yeah, let's let's continue on with this. but you don't want

Craig: it a docker container because that route, brings a whole lot of headaches. like, where you're saying, okay. docker container only has access to certain files that you give access to. if you want cloudbot to do other things, then you're really gonna have to go in that container and give access to all the things that it needs access to. it's probably okay for control, but it's probably a much bigger pain in the butt than you think. now i use that in docker all the time where i have, files below a certain directory that it can access, which i use to do a bunch of stuff with. so if your files exist below a certain directory, it's great. but, otherwise, you wanna give it access to your machine so it can spin up and do the things. otherwise, you're gonna run into, oh, that command doesn't exist in the structure that you gave it to me. i can't run those things. so you have some issues around that, so you gotta keep those in mind. but the idea of, you know, action over talk, i think what's really cool about cloudbot multbot is the stuff that people like myself and a bunch of other people way better than me have been doing for a long time now becomes more accessible. so instead of building these ai agents and taking the troubleshooting them and taking a bunch of time to build them, say, alright. now i've got this new ai agent that i'm gonna tweak over the next few weeks. somebody can now use this tool to do those things quicker and potentially better. so i and it like, it just makes everything more accessible. so you don't need to be a super expert in code for months to do the stuff that you can do with cloudbot within hours of install.

Jon: yeah. and for those of you that are like, well, why can't i just use cloud desktop? right? well, so it is multiple. it's you can use any llm, but, the thing is it's not just a chatbot. it's an autonomous agent. so, it integrates persistent memory, local execution, and workflow automation. so unlike typical chatbots that kind of reset context each session, motebot stores the history as local files. and so the more you use it, the more it understands you over time and can act on your behalf across different applications. so it is like a real person kind of thing as far as someone who can learn and anticipate your actions, can execute on command on things that you tell it to do. like, for example, you can tell it to, schedule events based on different contextual signals. you can automatically scan emails and update your crm. you can trigger jobs. you can interact with apps. you can do a whole bunch of things that someone, who is using a computer can do. right? so, it's really cool. i think, i love so, apparently, this guy has released 43 other different, projects, and this is the one that kind of took off. so, hopefully, he'll release something, or there'll there'll be others. there'll be other choices that come out, so just be on the watch out for sure. but this is exciting. this is where i like ai going.

Craig: yeah. you mentioned a couple of benefits like, inbox zero. your ai could filter, prioritize, and draft replies 90 of your emails before you wake up. your house can automatically adjust lights, temperature based on upcoming calendar events and the weather. autonomous planning, you could say, like, plan a birthday party for john, and ai would message, your friends on telegram or what have you, find a venue via google and put it on your calendar. so it can do a bunch of those things so that, you know, when people talk about an ai assistant, you know, up we're really going 2026 will be the year of ai agents even though people think 2025 was. it wasn't. we're now really starting to see agents start to take off and appearing everywhere. like i said, i build them all the time for different things. i've got a couple running in the background right now. and with products like cloud cowork coming out, which is going to do a lot of what you're suggesting here, but it's not, as easy as it is with mopbot. so as these products come out, motebot may end up being, you know, backbench as a result of people building these kind of tools into their products. but sam altman and many others say, you know, you've gotta, bet on us that we are gonna continually approve. and if you are trying to fill a minor gap where we have a deficiency, you know, the next revision, you're likely put out of business because we're gonna fill that gap. so you have to figure out complimentary access. and i think this is one of those things. all the high encoding tools, they're competing, you know, because the companies want this money themselves. they're burning so much cash. there were some articles recently about openai. you know? oh, they're will they go bankrupt? like, we've been talking about this for how long that their billions of dollars spend every month. they're farther and farther in the hole. you know, this happens. yeah. and that's why they have to keep going for fundraising because otherwise, they'll run out of money. and so they want those $100 subscriptions. they want those $200 subscriptions. they need that money to go. so they will start to take that business away from everyone else. that's why we're gonna start to see more advertising in these things. and i start to wonder, is advertising gonna change the world, from a ai perspective? will it be embraced or or not? i think it'll be embraced because we accept it, and that might actually start to make them profitable. so that'll be a different discussion. back to this thing, you know, it's not a releasable product. you gotta do some installs. you gotta muck around. so it's not quite ready to prime time. it does lower the barrier, but it's still not i can't just tell somebody to go install it. they're gonna have problems unless they got a clean mac, and that might work. it isn't cheap. you know? you need a dedicated computer generally running it, and potentially fifty to a hundred and fifty dollars a month in api fees for the ai to do its work and thinking time and whatnot unless you're powerful enough to run local ai. like you said, if someone hacks your local ai box, they have the keys to the kingdom, your email, your messages, your home security. ai still make mistakes and hallucinate, and ai accidentally canceling your flight because it misread an email is a real risk. so, you know, you wanna you have to give it guardrails. the system needs at least 16 gig, 32 is better. the system doesn't do everything locally. you use local orchestration with a cloud intelligence balance, privacy, and power. centralizing your entire digital life into one local bot, creates massive security vulnerability if not configured correctly, and it's a single point of failure. so the concept aligns with computer trends, in agentic workflow's self hosted and privacy focused design, making it plausible evolution of a personal ai technology. i think this is an alpha or beta version of what we're going to see as personal assistants. seen the rumors of openai is gonna launch headphones that, talk to a local ai and do stuff for you. like, they're all trying to be this local ai model and do what rabbit promised, like, last year or the year before, this action model. everybody's trying to make that leap, and we're seeing a thousand different, apps doing different things. so i'm really hyped to see a general agents coming that can do all these things because we're gonna see very much a more powerful bunch of agents that can do many more things. and now you're starting to realize why microsoft and apple are really worried about we're putting all these extra processors in for ai because you're gonna need them because you're gonna have all these agents running all the time. so you still wanna be able to use the computer when you wanna use the computer. you don't want them like, hey. sorry, man. we're busy right now. so, what i love this. again, i had some issues with compatibility with some other stuff i'm running with n a n and anti gravity and other things, so i have to take a step back and fix that this weekend. but it looks really cool, and i love, what it looks like you can automate with it. are you gonna be installing this soon, john?

Jon: i don't know. i don't know yet. i mean, i've, like, i've got 32 gigs of ram. i've got the m one max chip. i have to probably run it easily, but, i'm gonna do some more research on it and follow sort of the news. like, i've just there was just an article that just came out this morning that they discovered malicious, microsoft versus code extension that was designed for mold bot, and it was hosted on the official extension marketplace. so, there's a lot of, scare there's a lot of things that scare me about it. but, i don't know if we touched on this, but, cloud coworker is now available on the pro, subscription model. so, that's something i might look into as well? but, yeah, long short story long, i am going to at some point, but just not right away.

Craig: and what you mentioned about that versus code extension, it's probably what's happening with anti gravity because all those codes are shared by all low, versus code type ide interfaces, which anti gravity is one of them. and you can just go into the marketplace and grab them. so i'll mention that to my security person who is, on the case right now about what the sophos there is. but if that's the case, you know, it's so easy to get viruses or stuff running on your system because i probably have it running, didn't even know because it's a part of the official microsoft repository. let's go. you wouldn't think anything twice about it. and that's just an example of would you trust an ai to have the keys to your house and your inbox if it meant never having to schedule a meeting again? let us know in the comments if this sounds like a dream or a privacy nightmare, and we'll keep being a close eye on these local agents as they move from concept to reality.

More from Silicon Scoop

All episodes