Sideloading: Creating A Model of a Person via LLM with Very Large Prompt

Sideloading is the creation of a digital model of a person during their life via iterative improvements of this model based on the person's feedback. The progress of LLMs with large prompts allows the creation of very large, book-size prompts which describe a personality. We will call mind-models created via sideloading "sideloads"; they often look like chatbots, but they are more than that as they have other output channels, like internal thought streams and descriptions of actions.By arranging the predictive power of the facts about a person, we can write down the most important facts first. This allows us to get a high level of fidelity in a person's model by writing down a finite number of facts and running it as an LLM prompt with instructions to create a person's chatbot. This instruction also includes an extensive general sideload prompt (we will call it a prompt-loader), which explains how the person's sideload should work, but it is universal and works as an operating system for different persons.I created my personal sideload to test all this and it started working surprisingly quickly.We found that there are three main levels of facts about a person: core facts, long-term memory, and historical facts. Core facts should go to the main prompt of the LLM, long-term memory belongs to RAG, and historical facts – which a person often does not remember – should be only used to extract new relevant data. The most important thing in preparing data for sideloading is to get a full list of the core facts, and not to mix it with long-term memory and historical facts.The LLM-sideload project is inherently future-oriented. New state-of-the-art LLMs with large prompt capabilities emerge quarterly. This allows for the preparation of prompts that exceed current LLM capabilities, anticipating smooth execution in the following year.Below we present the key findings of this work in the form of an executive summary:Quality of sideloads is evaluated using three metrics:•Facts: Accuracy of the answers about the person's life•Vibe: How well the sideload captures the person's style and personality, as judged by the person themselves and their acquaintances•Brilliant insights: Ability to contribute unique, valuable ideas beyond the training data, while maintaining the author's style•Coarseness: Levels of details as percent of the total of recordable textual information from memory estimated to be 10000 pages. Similar to the jpeg level of compression. Estimated performance of Alexey Turchin's sideload:•Facts: 70% correct•Vibe: 20% accurate•Brilliant insights: near zero•Coarseness: 10%The project achieved these results in about one month of work. Vibe is harder to evaluate and regulate than facts. This approach focuses on informational personal identity, setting aside issues of observer identity and internal qualia.Improvement methods for sideloads include:•Iterative prompt updates based on feedback•Manual definition of vibe rules•Using LLMs to learn and refine vibe rules based on real-life communication chat logsAs of fall 2024, LLM-sideloads face several limitations:•Prompt size constraints•Limited availability of personal facts•Base LLM's knowledge and intelligence limitationsPros of sideloading for personal immortality:•Currently available, cost-effective, and rapidly improving•Enables long-term storage and multiple copies•Functions as personal assistant and memory aid•Offers potential applications in AI safety and cryonics•Does not require superintelligence for creation•Provides human-readable and customizable format•Allows privacy control over included informationCons of sideloading:•Lacks true consciousness/qualia (presumably)•Missing personal details and neural-level information•Challenges in creating high-quality versions•Risk of errors or unrealistic behavior•Ethical concerns (suffering potential, deactivation ethics)•Dependence on corporate-owned LLMs and cloud services•Privacy risks and potential data leaks•Possible influence on original person's behavior•Knowledge limited to creation date•Emotional and uncanny valley effects on users•Legal and social implications (consent, ownership, misuse)A sideload represents a distinct state of being – it is neither a copy, an upload, a chatbot, nor digital immortality.I have made my personality model publicly available through two channels:1.GitHub Repository:https://github.com/avturchin/minduploading/(Contains the personality data and prompt-loader versions)2.Interactive Sideload:https://chatgpt.com/g/g-EeFIesHsn-alexey-turchin-v7(Allows direct testing of the sideload implementation)3.Permanent address of this article https://philpapers.org/rec/TURSCA-3