The Geek Den

Blog

  • We’re Not Ready for Superintelligence

    Saw a great video on YouTube this morning, regarding humanity’s relationship with AI, and where things might be headed. Superintelligence will require a huge amount of responsibility from humans to be kept in check. Please take the time to watch and absorb this video: We’re Not Ready For Superintelligence.

  • My Journey with OpenWebUI

    July 4, 2025:

    Today, I am starting my journey using OpenWebUI, a self hosted AI platform.

    I will be adding to this post as I go along – sharing in the discovery, failures, learning, frustration, and successes of hosting my own local AI.

    July 16, 2025:

    Hosting your own AI has many benefits:

    Hosting your own local Artificial Intelligence (AI) solution can bring several benefits to individuals, organizations, and communities. Here are some advantages, according to my OpenWebUI instance using the LLM model Llama3.1(8B):

    1. Improved Data Privacy: By hosting AI locally, you maintain control over the data and ensure it remains within your premises or a secure, trusted location. This minimizes the risk of data breaches, cyber attacks, and unauthorized access.
    2. Enhanced Security: Local AI deployment reduces the attack surface for malicious actors to target, as there’s no need to transmit sensitive data to the cloud. You can implement robust security measures to protect your system from vulnerabilities.
    3. Increased Performance: Hosting AI locally allows for faster processing times and reduced latency, especially when dealing with large datasets or real-time applications. This is particularly beneficial in applications like autonomous vehicles, drones, or smart cities.
    4. Reduced Dependence on the Cloud: By having a local AI infrastructure, you’re less reliant on cloud services, which can be vulnerable to outages, downtime, or changes in their terms of service.
    5. Customization and Flexibility: Local AI hosting enables you to tailor the solution to your specific needs and requirements. You can modify the architecture, algorithms, and models to suit your organization’s unique circumstances.
    6. Cost Savings: While initial investment costs may be higher for local AI deployment, they can lead to long-term cost savings by reducing cloud service fees, data transmission costs, and potential penalties for non-compliance with regulations.
    7. Regulatory Compliance: By hosting AI locally, you may be better positioned to comply with specific regulatory requirements, such as data residency laws, GDPR, or CCPA, which mandate the storage of sensitive data within a country’s borders.
    8. Edge Computing Benefits: Local AI deployment enables edge computing applications, where processing happens closer to the source of the data (e.g., IoT devices). This reduces latency and improves real-time decision-making capabilities.
    9. Disaster Recovery and Business Continuity: With a local AI infrastructure in place, you can ensure business continuity during outages or disasters by switching to a secondary system or location.
    10. Research and Development Opportunities: Hosting local AI solutions creates opportunities for research and development (R&D) within your organization, allowing you to explore new applications, models, and techniques.

    Usefully Fast:

    Mind you, that response took only 13 seconds for my AI rig to take my prompt, which was “Benefits of hosting your own local AI”, to think about it, and then write this output for me (and you) in a clean, easy to read format. It also gave me “Follow up” questions like these:

    Follow up

    What are some common challenges or considerations when implementing a local AI solution?

    Can you provide more information on the hardware requirements for hosting a local AI solution?

    How do I choose the right AI framework or platform for my specific needs?

    Thoughts:

    As you can see, you get a lot of information and suggestions for follow up questions pretty quickly. That’s not too shabby for a 6 year old computer I cobbled together from spare parts, using a currently “normal” sized LLM. You can get information on an incredible range of subjects in a clear, concise output which you can use to help with your workflows and/or thought processes without spending too much money. From what I have learned, the most important part is to have a good GPU (Graphics Processing Unit) to do the work quickly, as AI speed is mostly determined by GPU, not CPU. Just to save you the time and hassle of doing the research yourself, here is an Amazon link to the GPU I’m using for my AI rig: NVIDIA 3060 12GB OC Edition. It is a cost-effective modestly priced GPU with quite a bit of VRAM, which is (primarily) what OpenWebUI uses to process your requests, and return the outputs.

  • New Page Added: Navigation

    Just added a Navigation page to the site to help Ya’ll find what you might be looking for. Here is a link: https://thegeekden.net/wordpress/navigation/. Please note that it is also available on the front/home/landing page for the site, in the top right corner, and located in the same place on all the other pages too.

  • Welcome to the Geek Den…

    Welcome to thegeekden.net! This is the WordPress Blog section of our site, where we will provide the stuff we want to post and/or share. We do this with the intent that some of you fine humans on the Internet will find something entertaining, informative, or useful for your own needs.

    If you are interested in 3D-Printing, DIY Projects, Linux, Open Source Software, Self-Hosting your own network services, Artificial Intelligence, Technology, you are in the right place.

    Please notice that we will provide informational links as much as possible in our posts to provide further context about the topics related to the current post. Just hover your mouse pointer over them to show you where they lead to, and click on the underlined words to navigate to them if you wish.

    Also, we will include affiliate links to products on Amazon, and other sites, so that we can (hopefully) make a little bit of money to help support our site, and to get our readers straight to the products we use ourselves for our projects.

    We will NEVER have advertisements on our website – Quite frankly, we find them distracting at best, and usually unproductive for what we are trying to accomplish, with regard to posting projects and information here on our site.

    Our ideology is in alignment with Tim Berners-Lee’s philosophy on the intent of the Internet: It should be a free, open, and decentralized “universal information system” that serves humanity, upholds democracy, and empowers individuals, which he famously summarized as being “for everyone”. He (and we) envision the web as a collaborative platform facilitating harmony, privacy, and equality. He is now advocating for a re-decentralized “pro-human” web where individuals own their data, a belief which we strongly agree with.

    The commercialization of the Internet by corporate interests is eroding user privacy, online autonomy, and the open structure of the Internet as a shared public resource. This shift, driven by “surveillance capitalism” and the dominance of a few large companies, has transformed the web from a decentralized network for information exchange into a platform designed for targeted advertising, data collection, and content manipulation.

    For more information on this, please see this article: “It’s not too late to fix it” at The Guardian.

    We will try to keep things as simple as is humanly possible in our posts, since it is getting harder and harder to keep up with the complexities of life in this modern world.

    So, without further adieu, onward, and upward!