HYPOTHETIC

hypothetic

Finding the right AI model to build a workflow around is hard. With so many models available across different platforms, it’s impossible to know where to start or how to find the one that best fits your specific needs. That’s the problem I set out to solve with AIModels.fyi, the only search engine for discovering and comparing AI models from across the web. The site has a simple premise: describe your problem in natural language, and get back a list of AI models that can potentially solve it. For example, you might search for “I need a model that can help upscale my images without losing clarity” or “make my cat look like it’s singing”. The goal is to , not just ML engineers, to find useful AI tools for their projects. make it easy for anyone Under the hood, the search works by taking the user’s query and comparing it to a database of over 240,000 AI models from 7,500+ creators across major platforms like Hugging Face, Replicate, Cerebrium, and DeepInfra. When a user types in a search, the query gets embedded into a vector using OpenAI’s text-embedding-ada-002 model. This query vector is then compared to pre-computed vector representations of the model names, descriptions, and use cases in the database. The most similar models by cosine distance are returned as the search results. Here’s a simplified version of the core search logic I’ve been using for about six months: const fetchData = async (query) => {
  // Embed the search query
  const embeddingResponse = await openAi.createEmbedding({
    model: "text-embedding-ada-002",
    input: query,
  });
  
  const embedding = embeddingResponse.data.data[0].embedding;

// Find the most similar models in the vector database 
  const { data: modelsData, error } = await supabase.rpc("search_models", {
    query_embedding: embedding,
    similarity_threshold: 0.75,
    match_count: 10,
  });
  return modelsData;
}; This vector search approach works reasonably well for surfacing relevant models when there’s a direct overlap between the query terms and the model descriptions. However, it struggles with more abstract, complex, or niche queries where the user’s language doesn’t exactly match the technical terminology in the database. For example, I recently wanted to find a model that could animate a portrait image using an audio voice clip to create a talking head effect. I knew there were some good models out there for this, but my searches for things like “talking portrait from image and audio” weren’t turning up the results I expected. The problem was that my query terms didn’t have enough literal matches with the specific words used in the model descriptions, even though the semantic intent was similar. This got me thinking about ways to improve the search experience to better handle those kinds of abstract or complex queries. How could I make the system smarter at understanding the user’s true intent and mapping it to relevant models in the database? That’s when I came across the idea of Hypothetical Document Embeddings (HyDE) from after reading about it in . this research paper this Medium post Meet Mr. HyDE — The query expander! The key insight behind HyDE is to use a large language model (LLM) to expand the user’s short query into a more detailed hypothetical “ideal” document before embedding it for search. So, instead of directly matching the query embedding to the database embeddings, it first translates the query into a richer representation in the same semantic space as the real documents. Here’s how the process works: The user enters a natural language query expressing their needs, like “I want to convert an image into a 3D model.” This generated pseudo-document is then embedded into a dense vector using a model like OpenAI’s text-embedding-ada-002. The resulting embedding is used to query the vector database and find the most similar real model documents. The query is fed into an instruction-tuned LLM like GPT-3 to generate a hypothetical document. The prompt asks the model to imagine an ideal document for answering the query. For example: Query: I want to convert an image into a 3D model Hypothetical Document: There are several AI models available that can convert a 2D image into a 3D model. The process is known as single-view reconstruction or 3D estimation. It typically involves using deep learning to predict the depth map and 3D shape from the 2D input image. Some popular models for this task include Pix2Vox, Pixel2Mesh, AtlasNet, and OccNet. These models are trained on large datasets of 3D objects and can handle complex shapes. To use them, you provide an input image and the model will output a 3D mesh or voxel representation that you can render or edit further in 3D modeling tools… The LLM step helps bridge the gap between the user’s simple query and the more technical language used in the model descriptions. It expands on the initial query to infer additional relevant terms, concepts, and context that can help retrieve better matches. At the same time, using real document embeddings as the final search index keeps the results grounded and prevents hallucination. I implemented a basic version of HyDE on top of the existing AIModels search, and the initial results are promising. In some sample queries I tested, HyDE surfaced relevant models that the original keyword-matching approach missed. For the talking head portrait query, it was able to generate a hypothetical document with key terms like “lip sync”, “facial animation”, and “audio-driven animation”, which helped retrieve some strong model matches that the plain search had overlooked. two to three times more How to implement HyDE in your own search system Let me show you how HyDE can be added in on top of an existing semantic search system to dramatically improve its quality. Let’s start with a simplified code sketch based on my current search flow. Here’s the full file — you can glance through it in totality before we look at the individual parts: const fetchData = async (query) => {
  // Generate a hypothetical document embedding for the query 
  const hypotheticalDocument = await generateHypotheticalDocument(query);
  
  const embeddingResponse = await openAi.createEmbedding({
    model: "text-embedding-ada-002",
    input: hypotheticalDocument,
  });
  
  const embedding = embeddingResponse.data.data[0].embedding;

// Search the real model embeddings using the hypothetical document embedding
  const { data: modelsData, error } = await supabase.rpc("search_models", {
    query_embedding: embedding,
    similarity_threshold: 0.75,
    match_count: 10,
  });
  return modelsData;
};
const generateHypotheticalDocument = async (query) => {
  // Use GPT-3 to expand the query into a hypothetical ideal document
  const prompt = `The user entered the following search query: 
  "${query}"
  
  Please generate a detailed hypothetical document that would be highly relevant to answering this query. The document should use more technical language and expand on the key aspects of the query.
  Hypothetical Document:`;
  const response = await openAi.createCompletion({
    model: "text-davinci-002",
    prompt: prompt,
    max_tokens: 200,
    n: 1,
    stop: null,
    temperature: 0.5,
  });
  const hypotheticalDocument = response.data.choices[0].text.trim();
  return hypotheticalDocument;
} Let’s see how it works. We’ll start with the function, which is the main entry point for the search process. First up: fetchData const hypotheticalDocument = await generateHypotheticalDocument(query); This line calls the function (which we'll explain in more detail later) to generate a hypothetical ideal document based on the user's search query. generateHypotheticalDocument The hypothetical document is a key part of the HyDE approach, as it aims to capture the user’s search intent in a more detailed and technical way than the original query. const embeddingResponse = await openAi.createEmbedding({
  model: "text-embedding-ada-002",
  input: hypotheticalDocument,
});
const embedding = embeddingResponse.data.data[0].embedding; These lines use OpenAI’s model to generate an embedding vector for the hypothetical document. I think it's the best embedding option out there so far. text-embedding-ada-002 Embeddings are a way to represent text as a dense vector of numbers, where similar texts have similar vectors. This allows us to perform semantic similarity searches. By embedding the hypothetical document instead of the original query, we aim to get a vector representation that better captures the expanded search intent. Next up: const { data: modelsData, error } = await supabase.rpc("search_models", {
  query_embedding: embedding,
  similarity_threshold: 0.75,
  match_count: 10,
}); This part performs the actual similarity search in the model database. It is called a remote procedure ( ) on the Supabase database, passing in the embedding of the hypothetical document as the . search_models query_embedding The and parameters control how closely the results need to match and how many to return. similarity_threshold match_count The search looks for model embeddings that are most similar to the query embedding based on cosine similarity. Now let’s look at the function: generateHypotheticalDocument const prompt = `The user entered the following search query: 
"${query}"

Please generate a detailed hypothetical document that would be highly relevant to answering this query. The document should use more technical language and expand on the key aspects of the query.
Hypothetical Document:`; This part constructs the prompt that will be sent to the language model to generate the hypothetical document. The prompt includes the user’s original query and instructions for the model to generate a detailed, technical document that expands on the query’s key aspects. Crafting an effective prompt is crucial for guiding the language model to generate useful hypothetical documents. Now for the next part: const response = await openAi.createCompletion({
  model: "text-davinci-002",
  prompt: prompt,
  max_tokens: 200,
  n: 1,
  stop: null,
  temperature: 0.5,
}); These lines use OpenAI’s model to generate the hypothetical document based on the prompt. text-davinci-002 The parameter limits the length of the generated document while controls the randomness (higher values make the output more diverse but potentially less focused). Adjusting these parameters can help tune the quality and variety of the generated documents. max_tokens temperature Next, we have: const hypotheticalDocument = response.data.choices[0].text.trim(); This line extracts the generated hypothetical document text from the GPT-3 API response. The generated text is then returned to be used in the embedding step. Then we can just retreive the data as before, this time using the hypothetical document as the query, but getting a lot more relevant results due to the expanded search term and all it’s extra, juicy context! Recap — what did we just build? I hope the code and my explanation demonstrate how the HyDE approach can be implemented on top of an existing semantic search system. It’s not that hard! The key idea is this: By inserting the hypothetical document generation and embedding steps into the search pipeline, we aim to improve the search results by better capturing the user’s intent. To recap, the key parts were: Using a language model to expand the user’s query into a more detailed, technical hypothetical document Embedding the hypothetical document to get a vector representation that aligns with the model database Performing a similarity search between the hypothetical document embedding and the model embeddings to find the most relevant results Of course, this is just a simplified implementation (simplementation?), and there are many details and optimizations to consider in a real implementation. But it illustrates the core flow and components of the HyDE approach. Taking it further Since this is just an initial prototype there’s still a lot of room for experimentation and refinement. Some of the key open questions I’m thinking about: What’s the optimal prompt structure and output length for generating useful hypothetical documents? I’d like to try different prompt templates and output sizes to see how they impact result quality. How can the hypothetical documents be made more interpretable and transparent to users? It could be confusing if users don’t understand why certain terms are being added. I’m considering showing parts of the generated text as a “query interpretation” to provide more insight into the process. To what extent does the result quality depend on the specific LLM and embedding models used? I’m curious to benchmark different model combinations (e.g. GPT-4 vs GPT-3, vs OpenAI embeddings, etc.) to find the best recipe. Contriever How can this approach be efficiently scaled to work over millions of database entries? Inference on large LLMs like GPT-3 is expensive and slow. I’ll need to look into caching, optimizing, or distilling the models to make this viable at scale. I’m excited about the potential for language models to help make search interfaces like AIModels more intuitive and powerful for a wider audience. Bridging the semantic gap between how users naturally express their needs and how AI models are technically described feels like an important piece of the puzzle. Techniques like HyDE point a way toward search experiences that deeply understand the intent and meet users where they are. My long-term goal is to keep refining this interface until anyone can easily find relevant AI models for their needs, no matter their level of technical expertise. I’ll be sharing more details as I continue experimenting with HyDE and other approaches. If you’re curious to learn more about the technical side, I recommend checking out the original paper. And if you have ideas or feedback on the AIModels search experience, I’d love to hear from you! Let me know what you think in the comments. Also published . here

Improve Semantic Search and Enhance AI Model Discovery with Hypothetical Document Embeddings (HyDE)

Ever stumbled upon an old, fuzzy snapshot of your ancestors and wished for it to be clearer? Or tried to zoom into a cherished picture from your buddy's Facebook collection, only to meet with unsightly pixelation? Don't despair. AI image super-resolution technology is here to infuse new vitality into your dull photographs. These models restore and enhance details in low-resolution pictures. They can even be used to clean up AI-generated images that have unsightly defects or artifacts. In this guide, we'll compare and contrast two ground-breaking AI models - and . Though they share similar names, they serve subtly different purposes. They both employ deep learning to augment resolution and correct degradation, but each excels in its own specific area: ESRGAN Real-ESRGAN : Offers outstanding results for ideal simulated degradation, such as bicubic downscaling. If you're dealing with pristine bicubic downscaled images or weird noise from AI generations, ESRGAN is still a good option for you, despite being older. ESRGAN : Tailored to tackle real-world corruptions like the compression common in social media. If your goal is to restore old photos or social media images affected by unknown blur or noise, Real-ESRGAN should be your pick. Real-ESRGAN also performs pretty well on synthetic images. Real-ESRGAN In this exploration, you'll understand how each model functions, its capabilities, limitations, and criteria for choosing one over the other. So, if you have a collection of vintage family photos needing an upgrade, or you desire to recapture the lost detail in your social media shares, or just want to clean up some Stable Diffusion generations, this guide is here to help you sift throught the the 'software magic' behind AI photo enhancement and place you with the right model. Subscribe or follow me on Twitter for more content like this! How ESRGAN brought a huge leap in image quality Back in 2018, a unique type of artificial intelligence model named ESRGAN revolutionized how we see and enhance images. It demonstrated the remarkable ability of AI to generate highly realistic textures and details for improving image resolution. This leap forward resulted in significantly better visual quality compared to older methods. How Does ESRGAN Work? At the core of ESRGAN, there's a friendly game of deception going on. One part of the system, called the 'generator,' tries to upgrade the resolution of a low-quality image. Meanwhile, another part named the 'discriminator' is trained to tell the difference between a high-quality image and one that's been artificially enhanced by the generator. The fascinating part? They're competing with each other. This competition pushes both parts of the system to get better and better, resulting in images that look astonishingly photorealistic. You can read more about this implementation in the . ESRGAN paper on Arxiv The Building Blocks of ESRGAN ESRGAN uses some cool techniques to produce the best possible results. Here's a quick look at them, in simpler terms: : These are the fundamental units of the 'generator' that help process and enhance features in the image. Residual-in-residual dense blocks : It's a kind of training method that ensures the system produces stable and consistent results. Relativistic GAN loss : This technique helps ESRGAN to recover textures in the image by comparing high-level image features from a pre-trained network. Perceptual loss However, there's one caveat with ESRGAN. It depends on a method called 'artificial bicubic downsampling' of high-resolution images during training. In layman's terms, this means it assumes a smooth and regular process of image degradation, which may not reflect the more random ways images can degrade in the real world. Nevertheless, ESRGAN was able to deliver top-tier image quality on research datasets and pioneered the use of AI for super-resolution, showing us a glimpse of the future of image enhancement. How Real-ESRGAN went from synthetic to real-world images ESRGAN is a star performer when it comes to enhancing 'clean,' digitally downscaled images. However, real photographs encounter numerous, often unpredictable distortions that can be tricky to replicate or correct. These can include: : Brought on by factors such as camera misfocus, object motion, or atmospheric disturbances. Blur : A result of imaging sensors or compression algorithms. Noise : Unwanted effects from JPEG compression or resizing operations. Artifacts : A combination of the above distortions, among others. Multiple degradation types To deal with these complex, real-world issues and enable super-resolution for actual photographs, Real-ESRGAN stepped up the game with some major enhancements: Key Improvements in Real-ESRGAN Unlike ESRGAN, Real-ESRGAN has a more complicated and robust system for restoring photos. Let's use some analogies to understand what each part of the algorithm is doing. The following descriptions are based on the , uploaded by the creator Xinntao. Note that you can also access the model if you'd like to further dive into the details. Real-ESRGAN paper on Arxiv on Github The Diverse Degradation Model: A Skilled Detective A good detective begins with understanding the problem. For Real-ESRGAN, the Diverse Degradation Model plays this detective role. It helps identify the many ways images can degrade - from noise to blur and compression artifacts. By understanding these real-world issues, Real-ESRGAN can plan an effective strategy to counteract them. This "detective work" allows Real-ESRGAN to tackle image restoration more effectively. The Sinc Filter: An Expert Artisan After understanding the problems, Real-ESRGAN acts as an artisan to start the restoration process. The Sinc Filter adds in subtle details, like the unique 'ringing' effects that are often present in high-quality, real-life photos. It's as if Real-ESRGAN uses the finest paintbrush to enhance the authenticity of your restored image. The U-Net Discriminator: The Quality Control Inspector Quality is key in any renovation. The U-Net Discriminator acts as Real-ESRGAN's internal quality control, analyzing every single pixel for perfection. This careful inspection ensures that the image restoration doesn't miss any fine textures and details, ensuring your photo's quality is top-notch. Spectral Normalization: The Project Manager Finally, to ensure a smooth renovation process, Real-ESRGAN needs a project manager. The Spectral Normalization feature keeps the training process balanced and stable, just like a project manager ensures a renovation project stays on track. This stability is crucial in achieving consistent, high-quality results. In essence, Real-ESRGAN combines these four elements to seamlessly transform low-quality, real-world images into high-resolution masterpieces. The Upshot: A Game Changer for Real-world Images These enhancements empower Real-ESRGAN to handle the challenging aspects of real-world image corruption with finesse. It greatly outshines ESRGAN when it comes to dealing with low-resolution images affected by unknown blur, noise, and JPEG compression artifacts. In other words, Real-ESRGAN is a game changer when you're trying to improve the quality of real-world images. Choosing the Right AI Tool: When to Use ESRGAN vs. Real-ESRGAN Artificial Intelligence is a powerful tool, but like any tool, its efficacy largely depends on the task at hand. Understanding when to employ ESRGAN versus Real-ESRGAN can make a significant difference in the quality of the outcome. ESRGAN: The Master of Synthetic Images ESRGAN is a virtuoso when dealing with artificially manipulated images. If you're working with images that have been changed by algorithms rather than natural processes, ESRGAN might be your best bet. For instance, consider digital artworks or images generated by other AI models. These synthetic images often exhibit a certain type of degradation, like bicubic downscaling, which ESRGAN is specifically designed to handle. It's the perfect tool to refine the textures and details of these images and give them a photorealistic touch. Real-ESRGAN: Conqueror of Real-World Corruptions On the other hand, Real-ESRGAN is a powerhouse for real-world photographs. The unpredictable nature of real-world corruptions, such as blurs, noise, or compression artifacts, can be quite a challenge to rectify. But with Real-ESRGAN's diverse degradation model and other advanced features, these challenges become manageable. Whether it's enhancing old family photos, improving social media images, or providing clarity to medical or satellite images, Real-ESRGAN is a real-world savior. Which AI model is better? The choice between ESRGAN and Real-ESRGAN is not a matter of which one is "better". Rather, it's about finding the best fit for your specific needs. While both models excel at enhancing images, they shine in different scenarios. By understanding the unique strengths of ESRGAN and Real-ESRGAN, you can effectively leverage these powerful AI tools to maximize their potential. Whether you are working with synthetic images or real-world photos, the right AI can make all the difference. Ideal Scenarios for Each Model : If your images have undergone bicubic downsampling, ESRGAN still achieves excellent results. It's built to handle this kind of synthetic degradation effectively. Bicubic Downsampling : For images distorted by real-world elements like camera blur, sensor noise, or compression artifacts, Real-ESRGAN is a much stronger choice. It's designed to handle these unpredictable corruptions that ESRGAN isn't ideally suited for. Real Photographs with Complex Distortions : Another advantage of Real-ESRGAN is that it can be used with GFPGAN, which enhances its ability to restore faces in photographs. If you're trying to improve the clarity of real facial images, Real-ESRGAN has an edge. Face Restoration Recognizing Their Limitations While both models are impressive, they also have certain limitations that are important to acknowledge: : If an image is of extremely low resolution or has been heavily degraded, both models may struggle to produce satisfactory results. The lack of details in the source image can limit what the models can recreate. Extremely Low Resolutions or Heavy Degradations : Neither model is ideally suited for video super-resolution due to flicker effects. As they enhance each frame individually, inconsistencies can occur, causing a flickering or jumpy appearance in the video output. Video Super-Resolution : Both ESRGAN and Real-ESRGAN require powerful GPUs and significant memory. If your hardware isn't up to the task, you might experience performance issues. Hardware Requirements By keeping these guidelines in mind, you can make an informed decision about which model to use for your specific needs, and get the most out of these powerful AI tools. The Best Upscaler AI Models: Exploring Alternatives When selecting an AI model for your project, considering different options can help you make a more informed decision. Here are a few AI models that offer similar functionalities to the ESRGAN and Real-ESRGAN models: by Tencentarc: This model specializes in face restoration for old photos or AI-generated faces. It's comparable to ESRGAN in its image enhancement capabilities but specifically . It operates on the Replicate platform under the Image-to-Image category and costs $0.0033 per run. Gfpgan targets facial images by Sczhou: Similar to ESRGAN, . The main contrast lies in the specific restoration algorithms utilized. Running on Replicate, it falls under the Image-to-Image category and costs $0.0055 per run. Codeformer Codeformer provides a robust solution for face restoration in old photos or AI-generated faces You can check out this guide to compare GFPGAN and Codeformer. by Jingyunliang: This model provides image restoration capabilities using the Swin Transformer. While ESRGAN and Real-ESRGAN use GAN-based approaches, for image enhancement. It operates on the Replicate platform, under the Image-to-Image category, and costs $0.0276 per run. Swinir Swinir's primary focus is on leveraging transformer architecture by Cszn: Scunet, a practical model for blind denoising, also has a notable contrast in methodology, utilizing the Swin-Conv-UNet and Data Synthesis. It runs on Replicate, classified as Image-to-Image, and costs $0.00275 per run. . Scunet Further reading on this model is available here These models provide varying solutions for image restoration and enhancement tasks. Though they serve similar purposes, the differences in their design, methodology, and cost per run demonstrate the variety of options available for developers. When selecting a model, consider your project's specific requirements, your available budget, and the unique capabilities and methodologies of each model. Conclusion Image enhancement has seen a significant evolution with the advent of AI models. They can breathe new life into your cherished but dull photos, enhancing details, and correcting degradation. Both ESRGAN and Real-ESRGAN utilize AI to serve this purpose, but each has its unique forte and is suited for different scenarios. ESRGAN shines when enhancing bicubic downscaled images, while Real-ESRGAN excels in managing real-world corruptions like blur, noise, and compression artifacts. By comprehending the unique strengths and limitations of ESRGAN and Real-ESRGAN, you can effectively choose the right tool for your specific needs. Whether it's enhancing synthetic images or real-world photos, the right AI can be a game-changer. However, it's crucial to understand that both models have their limitations, particularly with extremely low-resolution images or those with heavy degradation, and require powerful hardware for optimal performance. Remember, the choice between ESRGAN and Real-ESRGAN isn't about which one is superior overall, but which one is best suited for your unique needs. Thanks for reading and happy enhancing! Subscribe or follow me on Twitter for more content like this! Additional Resources Here are some useful resources to further explore the ESRGAN and Real-ESRGAN models, as well as their implementations: : The academic paper presents the Real-ESRGAN model, discussing its design, methodology, and results in-depth. Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data : This academic paper outlines the conception, design, and capabilities of the ESRGAN model. ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks : The detailed page for the ESRGAN model, including its description, specifications, capabilities, and cost per run. ESRGAN details : A popular Real-ESRGAN implementation available on the Replicate platform, created by NightmareAI. Real-ESRGAN implementation by NightmareAI : The implementation of Real-ESRGAN by the original author, Xinntao, also available on the Replicate platform. Real-ESRGAN implementation by Xinntao These resources provide comprehensive insights into the design and functionalities of the ESRGAN and Real-ESRGAN models, as well as their practical applications. They can be beneficial for anyone interested in these models, whether you're a developer looking to integrate them into a project or just someone interested in learning more about cutting-edge AI technology.

ESRGAN Vs. Real-ESRGAN - From Hypothetical to Real-world Super-resolution With AI

I recently interviewed for a company in a completely different sector this week. As I feared, I am a little too senior but it made me think — ‘what could I do bigger and better?’ So I spent the following morning writing up a strategy for a green economy startup that would use generative ai large language models, , and good old-fashioned hard work to tackle the challenge of climate change. web3 The result is ‘ ’ — an ambitious plan to acquire 1 million hectares of land and remove 1 billion tons of CO₂ emissions by 2050. Planetary How? trained on peer-reviewed, private and publicly available research on climate change, geospatial and satellite data and economic reports. The development of a large language model ( ) and proprietary machine learning algorithms LLM in order to create carbon credits in areas that will become increasingly more valuable as climate change takes hold. The acquisition of land identified using the proprietary AI platform that will be regenerated of events and transactions that impact each parcel of land acquired and trading platform for carbon credits. The development of a carbon-negative Layer 1 blockchain ledger that acts as a publicly available and transparent record Of course, this is all hypothetical and why would I literally write up a startup strategy to give it away for free? Partially, because ideas on their own are worthless these days, partially to show that many don’t think big enough, and partially to show that technologies can be used in the right combination to effect positive change that isn’t just more of the same. If people like Adam Neumann can secure then why is it that startups delivering sustainability and ESG solutions really struggle to capture the same attention of venture capitalists and investors? $350m for WeWork 2.0 I hope you enjoy reading this. Feel free to comment or pop me an email if you want to discuss some of the ideas. (and yes, I really did write this up in the morning after mulling it over in the bath the night before lol) Content Overview Introducing Planetary: A Hypothetical Company Acquiring 1M Hectares and Removing 1B Tons of CO₂ by 2050 Planetary’s Strategy Planetary’s Technology Profile The People of Planetary Introducing Planetary: A Hypothetical Company Acquiring 1M Hectares and Removing 1B Tons of CO₂ by 2050 is the biggest threat to human civilization and life on Earth we face today. Over the next century, we are set to face increasingly hostile environmental impacts on society and the ecosystem on which we depend. Increasing temperatures, rising sea levels, extinction of native flora and fauna, and the migration of species and humans from lands unfit for habitation to other areas less affected will create economic and societal imbalances we are not prepared for. Climate change But there is still time to act. And it is now. Our goal is to acquire 1 million hectares of land to remove over 1 billion tons of CO2 from the atmosphere by 2050. The immediate response is ‘ ’ given the scale needed by a single organization to achieve this, but we are committed to making it happen through a business model that not only has a positive and regenerative environmental impact but also provides a naturally sustainable and future-focused alternative that can benefit the many. impossible This is saving civilization on a PLANETARY scale. Planetary’s Strategy We intend to introduce a three-stage plan for the regeneration of land and sustained removal of carbon dioxide. Our vision is to become not only a leader in affirmative action against climate change but to leave a lasting legacy of investment for generations to follow. In order to support this, we will focus on three key areas, all of which are revenue-generating with the intent to increase in value as we scale. Our ethos for planetary transformation is People, Planet and Technology. Our immediate differentiators (that currently do not exist in full) within competitor strategy are as follows: trained on peer-reviewed, private, and publicly available research on climate change, geospatial and satellite data, and economic reports. The development of a large language model (LLM) and proprietary machine learning algorithms that will be regenerated in order to create carbon credits in areas that will become increasingly more valuable as climate change takes hold. The acquisition of land identified using our proprietary AI platform that acts as a publicly available and transparent record of events and transactions that impact each parcel of land acquired and trading platform for carbon credits. The development of a carbon-negative Layer 1 blockchain ledger Planetary’s Technology Profile it is necessary to develop a proprietary research platform dedicated to the identification of land that can be used for sustainable regeneration and growth whilst In order to acquire 1 million hectares and maximize the immediate and future return on value to investors and the amount of carbon credits that can be issued, protecting that investment against the impact of climate change over a longer period of time. As part of this effort, . By analyzing datasets like soil conditions, rainfall, and species suitability, we will create internal high-precision predictive machine learning models to inform our land acquisition strategy worldwide with a view towards long-term value and scalable carbon offset returns. It is especially we will develop open large language research models for licensing to optimize reforestation planning, sustainable development and farming, ecotourism opportunities, and more important to stress that a land acquisition strategy should not only focus on what is available today but what will be valuable for sustainable development beyond 2050 as the climate impact is felt. In this regard, it is important to take into account factors like wildlife and human migration, temperature fluctuations, crop degradation, and rising sea levels. The algorithms will analyze available data including soil conditions, rainfall, temperature, biodiversity, and invasive species to determine the most suitable types of projects. The same platform will help us identify and optimize the best regeneration efforts for maximum return. We will distribute these models with a permissive open source-software license which will be the first of its kind in this category. We will look to source a core research team not only of exceptional developers but of top planetary science researchers to continually update and validate input into these models. The models will be continually updated to reflect real-time and real-world data captured under each land parcel acquired and regenerated, providing confidence to investors and adding . further validation to the carbon credit marketplace To this end, we will develop a Layer 1 blockchain-based platform for the purchasing and acquisition of land and to openly track the progress of regenerative projects, carbon emission reduction, and fractionalized sale. In order to commit to radical transparency and ethics oversight it is our belief that decentralized technologies hold the key. Each parcel of land acquired becomes its own Layer 2 ledger capable of recording a public and transparent history of positive change. . The credits will represent the tons of carbon dioxide removed from the atmosphere through conservation projects thus creating transparent progress towards achieving our goal of removing 1 billion tons of CO2 by 2050. In addition, using smart contracts for the purchase of carbon credits allows us to set a percentage ‘royalty’ on secondary markets by a number of trades until retired. This Layer 2 will also act as a ledger system for the trade and resale of carbon credits issued against each project per plot of land The proprietary machine learning platform will be used to analyze satellite imagery and sensor data to accurately measure and verify the carbon impacts of different projects. Both platforms will also be used to monitor available real-time data and satellite imagery . The supply chain data would be recorded on for transparency and further proof of regenerative activities and carbon dioxide removal. to track the provenance of agricultural and forestry products back to verified sustainable sources Layer 2 Using blockchain also allows for future tokenization and empowered community oversight opportunities of the acquired land. This is explained later in the business model. To achieve a carbon-negative blockchain platform we are evaluating a number of existing solutions that can complement our own strategy. It also has a connecting layer, or “bridge,” that allows value and data to be transferred between most blockchains — and can even be used to connect to non-blockchain databases. Polkadot’s configuration of mainnet, parachain, and substrate layers fits with the primary business model and land-holding ownership structure of Planetary and can support tokenization and fractionalized governance. The Polkadot network includes a main blockchain called the “relay chain” and many user-created parallel chains (or “parachains”). Polkadot, Avalanche and Cosmos are within the same goals we wish to achieve with our platform. It has a specific goal to be as carbon neutral as possible by working closely in partnership with existing carbon reduction projects. It was specifically developed for web3 video game projects however many of the features in terms of tokenization of assets and Guild support are in line with our own plans. Confirmation from their CTO has suggested using their chain could be suitable for Planetary and the existing focus on the support of regenerative projects could become a beneficial partnership also. WAX is a third-generation blockchain based on a Delegated Proof of Stake consensus algorithm (DPoS). It is the first cryptocurrency to use a directed acyclic graph data structure, by having a “block” consisting of only one transaction and the account’s current balance. It is possible to continue to develop using Nano’s data structure and Proof of Stake consensus as a means of recording all activity within each land parcel acquired. Nano uses a block-lattice data structure, where every account has its own blockchain for storing transactions. Instrumental to this planetary transformation strategy and to create an exponential return not only for the investment in the enabling technology but the future of the sustainable removal of 1 billion tons of CO2 by 2050 is the acquisition of 1 million hectares under active management as a valuable and appreciating asset. The strategy would focus on using the AI platform to identify and acquire degraded or threatened habitats including forests, wetlands, existing farmlands and grasslands in key wildlife corridors and regenerate the ecosystems through ecological restoration and sustainable land management with a localized boots-on-the-ground approach. We would then implement biodiversity monitoring systems to quantify improvements in species populations, soil health, carbon sequestration, water purification, etc. These become verified impact metrics recorded on chain in our blockchain platform. Developing sustainable nature-based enterprises on the land generating eco-certification premiums — such as organic agriculture, sustainable forestry, eco-tourism, and renewable energy — allows for further reinvestment of profits into maintaining active conservation of the acquired land. Issuing conservation and carbon credits backed by the measured natural capital assets and ecosystem service benefits created on the land generates revenue and protects the capital asset value. This would be backed by using our proprietary algorithms to verify results and recorded on chain. We would actively explore pay-for-success models where governments or insurers pay for demonstrated social and environmental outcomes like flood protection, water filtration, or fire risk reduction. . This not only generates an immediate return on every capital acquisition but allows the local and global communities the opportunity to purchase a share in an appreciating asset and stake agency in active conservation at a price point that is affordable. As regeneration projects typically take up to 10 years or more to allow an organization or project to issue carbon credits we intend to fractionalize 30% of every land parcel acquired for purchase For illustrative purposes, we would buy 100,000 hectares of land at $25m and plan to sell 30% as fractional ownership but recover $50m as a result of the sale. Total land area: 100,000 hectares Fractional ownership to be sold: 30% of 100,000 hectares = 30,000 ownership shares Total amount to recover: $50,000,000 To recover $50,000,000 from the sale of 30,000 fractional ownership shares: $50,000,000 / 30,000 shares = $1,667 per share The fractional sale of land assets would not only generate an immediate return on the capital investment but additional funding to be placed in a separate foundation that would create the regeneration projects and invest in green startups. The added benefit of using decentralized technology is that every share would be executed as a smart contract on the Layer 2 solution. By creating a tokenized system it would allow greater decentralized control over the operation and avoid potential IPO under traditional structures should this be the consensus. There would be future opportunities to tokenize each share and create decentralized autonomous organizations per land asset acquired. The overall approach creates a positive feedback cycle where planetary conservation generates returns via natural capital assets appreciating in value as the land restores and institutional and retail investors are rewarded whilst enabling the work to pay for itself over time by aligning incentives through carbon sequestration and sustainable business. The People of Planetary Years ago that I heard about a man named Brendon Grimshaw that changed my perspective on the effect that one person with a mission can have on conservation. He bought an island in Seychelles in 1962 and for 50 years spent the remainder of his life along with his own ‘Man Friday’ to re-introduce the island’s natural flora and fauna that had been destroyed through tourism and trade. He planted trees, reintroduced bird species and the Aldebra tortoise that had become extinct across a number of the surrounding islands. He refused offers to buy the island and to turn it into another tourist trap. and is a story that should be told and taught to current and future generations that the natural environment can be regenerated if we dare ourselves to work with nature, ourselves and have the patience for it. The type of thoughtful conservation, regeneration, and rewilding of wildlife and the natural environment was achieved by a singular, committed vision to achieve and deliver a vision like Brendon’s but on a scale that can have not only a massive positive impact on the planet but on people, using technology. It is the same passion that drives the founding team at Planetary People, Planet, and Technology in harmony for environmental transformation. We intend to live to the ethics and vision we are starting and will minimize our own carbon footprint as much as possible. The founding team comprises of seasoned and experienced professionals with technology, green and economic backgrounds, operating remotely. Theo Priestley An Economist A Green industry analyst A PhD. in Earth Observation A senior Blockchain engineer The first five employees that have already been identified will be a mixture of experienced researchers from existing greentech and cleantech companies and developers who are motivated by the scale and vision to create a lasting legacy and impact on not only the planet and its living ecosystem but the people also. Further roles to be considered for early hire are land manager, ecologist, spatial data analyst, planning and property, legal, and finance. Some of these positions will be hired on a needs basis and deployed per land asset acquired to manage the estate and engage with the communities directly. Also published here.

Web3 & AI: A Hypothetical Company's Plan to Remove 1B Tons of CO₂ by 2050

Facebook’s newly forged blockchain group represents a massive strategic decision for Facebook, like David Marcus (previously Head of Messenger and Paypal President) and Kevin Weil (previously VP Product, Instagram and Twitter) placed at the helm. This piece explores the 4-part laddering up strategy Facebook could take to (1) go from P2P payments (2) to B2C within FB (3) to B2C across the web/mobile (4) to finally integrating with offline retail — all with dramatically lower processing fees to motivate merchant adoption. with top executives One must wonder how far behind Google, Apple and Amazon might be. Facebook’s recent restructuring represents the first strategic declaration of intent in crypto from a FANG incumbent. You know what’s cool? A billion+ users to stabilize a new cryptocurrency. Despite the explosion of ICOs the the past few years, They’re all missing two core ingredients: almost every cryptocurrency is sub-scale to serve as a truly stable medium of exchange. A massive, engaged end-user base use cases High frequency That’s where Facebook’s 2bn+ active users and B2C/P2P payment use cases — in Messenger/WhatsApp, as well as commerce (ads) — come in. , Amazon, and Apple have their own analogous positioning, too. Google This comes at a time that’s been particularly choppy for — amidst the l around user data privacy and , Zuckerberg has been on a world tour of . Facebook Cambridge Analytica scanda #fakenews election fallout contrition But advertising revenue relies on that data, and Facebook has struggled to shift their monetization model elsewhere. They’ve poured resources into Messenger over the past years, trying everything from and partnerships like , but nothing has gained traction at scale. customer service bots, shopping Uber Back in January, Zuckerberg first hinted at blockchain as a part of their long-term strategy: “Many people now believe technology only centralizes power rather than decentralizes it. There are important counter-trends to this — like encryption and — that take power from centralized systems and put it back into people’s hands. But they come with the risk of being harder to control. I’m interested to .” cryptocurrency go deeper and study the positive and negative aspects of these technologies, and how best to use them in our services Now, between of its decentralized nature and efficient payment model, could building a “ZuckCoin” on the blockchain help Facebook counter data privacy headwinds develop a new monetization model? and Facebook’s strategy for Crypto “Ladder-Up” “This is how a company accomplishes what, at the beginning, may seem impossible: a series of steps from here to there that build on each other. Moreover, it is not only an impressive accomplishment, it is also a powerful moat; whoever wishes to compete has to follow the same time-consuming process.” — Stratechery Facebook has a clear path to integrating crypto across its product offerings, each step building on the last with two-sided network effects between end-users and merchants in the process: Give a reason to use crypto 2bn+ people Lower processing fees to motivate merchant adoption Move from “Sign in with FB” to “Pay with FB” Enable , following WeChat’s playbook offline, low-fee payments Rung 1: Give 2bn+ people a reason to use crypto Facebook has + users across FB, Messenger, WhatsApp, Instagram — coincidentally, these form the recently restructured “ ” group under Chris Cox. That compares to of crypto user globally. 2.2bn Family of Apps tens of millions Facebook’s reach is 20x+ higher than crypto’s. daily First, ZuckCoin becomes a new way to send P2P payments to any connection. Versus traditional payments, ZuckCoin processes with effectively no fees and faster clearing times. Suddenly, users begin to see their network as functional connections — they can remit money home on Messenger and pay friends for dinner on WhatsApp. For the 1bn+ DAUs that live outside the US and Europe — over 67% of FB’s users — than their local, more volatile currencies. storing money in a ZuckCoin wallet could also be a more stable store of value Rung 2: Lower processing fees to motivate merchant adoption “ . So I think what you’re going to start to see are people interacting with pages, maybe follow a page on Facebook or Instagram. You see content from that page. You can click through or tap through to a message thread, and then you can either get customer support or complete a transaction or do a follow-on transaction.” — Messaging can be a more transactional medium than feed Zuckerberg Next, those same users start to possess more ZuckCoin. They have the opportunity to liquidate those tokens into fiat, leveraging Facebook’s existing global payment network. This network, bolstered by their ads and marketplace businesses, can accept payment in over , and partners with to facilitate global pay out rails. But with enough native FB use cases, there’s less reason to transfer into fiat. 50 currencies Transferwise For example, a user could click through a targeted ad directly to purchase and pay with one easy button —immediately delivering a superior user experience and clearer merchant attribution. With Facebook’s 80m+ businesses across the world, it’s easy to imagine organic transaction opportunities across ad units, customer service touch points, and brand pages. Why would the merchants want that? Well, for starters, e-commerce businesses already operate on incredibly . For companies that are striving to hit single-digit profit margins, are material. A blockchain-based coin could lower fees via Facebook to basis points, lean margins 2–4% payment processing fees giving the merchants full percentage points of profit back. Those retailers are suddenly incentivized to promote ZuckCoin as their preferred payment method — they can offer bigger discounts and loyalty rewards to that audience. With a clear margin motivation, merchants actually become ZuckCoin’s biggest advocates. Rung 3: Move from “Sign in with FB” to “Pay with FB” Today, “Sign in with FB” is by far the most common account sign-up and log-in flow. Tomorrow, . “Pay with FB” could become the most common payment flow, piggy-backing on top of the existing FB authentication widgets across the web/mobile Online merchants care deeply about 3 problems: Basket rate abandonment Payment fees processing charge rates Fraudulent They want to minimize all three. Paying with FB should, in theory, address that: Since sign-in with FB is already 1-click and 1.5bn+ users would potentially have stored ZuckCoin credentials, it would be much than manually entering shipping and CC info. faster to complete a purchase As mentioned in the previous section, . processing fees could drop by an order of magnitude Because Facebook has logged-in users with rich identities and social graphs, it should have the data to decrease fraudulent charges. Consumers crave lower friction payments, merchants want higher margins and less fraud, and FB can provide both across the web and mobile apps — making for a massive potential payment opportunity. Rung 4: Enable offline, low-fee payments, following WeChat’s playbook Lower fees are equally appealing to merchants offline, which still account for spend. 90%+ of retail Similar to ’s approach in China, For the 2bn+ people with a FB-owned app on their phone, they could simply open that to get their profile (QR)code to pay at checkout. WeChat Facebook could build POS-integrations for “Pay with Facebook” in the ROW. Imagine how transformative this could be — massive retail industries like getting 200bps of processing fees back. groceries, which have a 1.6% net margin , could more than double their net revenue, Beyond Merchant Payments These rungs simply describe the natural extension of a blockchain-based Facebook payment token to P2P and B2C transactions. ZuckCoin could also enable a whole new economy of that would never make sense in today’s world of high fixed cost processing fees. These micropayments would naturally support Facebook’s existing news and video content as a monetization model alternative to advertising. They could also be used to curate content and foster a thriving Facebook-native media marketplace. micropayments Finally, Facebook could decouple parts of , allowing users to own their digital data more directly. In that scenario, FB personal data generated would be encrypted, and users would hold their personal data decryption key. They would have control over sharing their data, with advertisers, 3rd party apps, etc and revoke access at will. While this in some ways would undermine Facebook’s own data moat, it . identity may be better for Zuckerberg to regulate his own company through technology rather than let Washington do it through policy “It’s inevitable that there will need to be some regulation, [but] you have to be careful about what regulations you put in place.” — Mark Zuckerberg, Apr 2018, Congressional Testimony

Facebook’s (hypothetical) crypto strategy

The Case for Software Supply Chain Security. A couple of days ago, on Oct 22, 2024, the Securities and Exchange Commission fined millions of dollars for mislabeling the risks of the SolarWinds attack as “hypothetical.” Four companies Unisys, Avaya, Check Point, and Mimecast were fined $7 million. In 2020, the SolarWinds attack affected major organizations when malware gave backdoor access to its Orion software. Supply chain attacks have only grown in both quantity and magnitude since then. Though companies continue to downplay the impact of Supply Chain security, on average, less than 50% of the code base in your organization is written by your employees. Supply chain security controls mitigate risks coming from the “external” software that ships with your product. The Cure to Make Robust Software Supply Chains. Google has invested heavily in insider risk since 2010 and has been building multi-layered security controls to defend against advanced attacks. Around 2019/20, Google Chrome started developing a framework that applies to a broader community, starting with Chromium. Collaboration across the company led to the formation of the supply chain levels for software artifacts aka SLSA. Since then, Google has donated this project to OpenSSF, one of the largest open-source software foundations. SLSA version 1.0 was launched earlier this year and it provides an incremental solution to gradually improve security posture from development to deployment for the software supply chain. SLSA Threat Model Let’s take a look into how malicious actors exploit weaknesses throughout the process right from build systems to the artifact distribution level. The SLSA threat model states that issues like inadequate code review, misconfigured build systems, and unverified third-party code-introduced risks, etc. can cause significant security breaches. The successful attacks on SolarWinds where the hackers injected malicious code into the build process require strong controls. Knowing your dependencies and preventing unilateral access remains crucial for ensuring malware can’t be inserted. Applicability of SLSA in Real World Scenarios There are three major use cases for SLSA, First Party, Open Source Projects, and Vendors. In First Party, since there is no need to transfer trust across organizational boundaries, this makes it the easiest use case. SLSA can be used entirely within an organization. In the case of Open Source Projects, SLSA uses a canonical source repository approach where users of open-source software need to only trust a handful of build platforms instead of trusting thousands of developers who have upload permissions to many packages. SLSA would simply reject an upload if it was not from a canonical source repository. And finally, for users of vendor-provided software, SLSA would require a vendor to be certified by a third-party auditor. This would make it easier for consumers to trust vendor software products. Also, provenance is the main aspect of using SLSA, it would provide traceability and verifiability of where the software originated. The goal is not just to prevent tampering but also to trace who was involved and where the exact origin of the software was. Since it facilitates transparency, companies can easily adapt to regulatory restrictions while ensuring that their business operations are sound. Hardening the Chain of Trust The core of SLSA is provenance, which documents the origin and security of artifacts. SLSA can pinpoint tampering or any other malicious behavior using cryptographic validation and automated attestation. Since it verifies that each step from creation to distribution of the software is indeed secure, the chain of trust is therefore solidified. The majority of well-secured development environments apply the principles of provenance including Google. One of the innovators of the SLSA framework at Google is Akash Mukherjee. He discusses how Chrome ensures verifiability in its builds in this article Chromium Chronicle #23. The article focuses on the fundamentals of the framework and the four pillars that we see today in SLSA. Meeting Government Mandates With SLSA Another advantage of SLSA compared to other security libraries is its ability to meet U.S. government regulations. The National Institute Of Standards and Technology aka NIST plays a crucial role in cybersecurity by creating frameworks, guidelines, etc. like the Secure Software Development Framework (SSDF). SLSA is able to help companies meet these requirements. Following this framework will also align companies with the Executive Order 14028 directive which has a higher standard for cybersecurity along the entire software supply chain. Also in highly regulated defense, healthcare, and finance industries, SLSA is reliable enough to be the security blueprint for organizations that can help generate evidence of compliance even when it touches critical cybersecurity concerns. The Future of SLSA as a Blueprint for Secure Software Development But this journey does not stop here; it continues evolving the industry. With the flexibility in its structure, SLSA allows organizations to enhance their security step by step while preparing for future threats. As SBOMs become central to software development, experts focus on securing development practices in general, “SBOMs can help lift security health from the status quo, but frameworks like Supply-chain Levels for Software Artifacts (SLSA) and Secure Software Development Framework (SSDF) must be used to improve security as you’re building new packages.”, Akash Mukherjee, an early developer of the SLSA framework in his recent post on RSA conference. By integrating the best practices from SLSA with development insights gleaned from resources such as Chromium Chronicle, an organization is well-equipped to design secure software. At its core, SLSA revolves around provenance, providing cryptographic assurances and compliance with regulatory requirements, making it an extremely important tool for securing the global software ecosystem. The Case for Software Supply Chain Security. A couple of days ago, on Oct 22, 2024, the Securities and Exchange Commission fined millions of dollars for mislabeling the risks of the SolarWinds attack as “hypothetical.” Four companies Unisys, Avaya, Check Point, and Mimecast were fined $7 million. In 2020, the SolarWinds attack affected major organizations when malware gave backdoor access to its Orion software. Supply chain attacks have only grown in both quantity and magnitude since then. Though companies continue to downplay the impact of Supply Chain security, on average, less than 50% of the code base in your organization is written by your employees. Supply chain security controls mitigate risks coming from the “external” software that ships with your product. The Cure to Make Robust Software Supply Chains. The Cure to Make Robust Software Supply Chains. Google has invested heavily in insider risk since 2010 and has been building multi-layered security controls to defend against advanced attacks. Around 2019/20, Google Chrome started developing a framework that applies to a broader community, starting with Chromium. Collaboration across the company led to the formation of the supply chain levels for software artifacts aka SLSA. Since then, Google has donated this project to OpenSSF, one of the largest open-source software foundations. SLSA version 1.0 was launched earlier this year and it provides an incremental solution to gradually improve security posture from development to deployment for the software supply chain. SLSA version 1.0 SLSA version 1.0 SLSA Threat Model SLSA Threat Model Let’s take a look into how malicious actors exploit weaknesses throughout the process right from build systems to the artifact distribution level. The SLSA threat model states that issues like inadequate code review, misconfigured build systems, and unverified third-party code-introduced risks, etc. can cause significant security breaches. SLSA threat model SLSA threat model The successful attacks on SolarWinds where the hackers injected malicious code into the build process require strong controls. Knowing your dependencies and preventing unilateral access remains crucial for ensuring malware can’t be inserted. Applicability of SLSA in Real World Scenarios Applicability of SLSA in Real World Scenarios There are three major use cases for SLSA, First Party, Open Source Projects, and Vendors. In First Party, since there is no need to transfer trust across organizational boundaries, this makes it the easiest use case. SLSA can be used entirely within an organization. use cases use cases In the case of Open Source Projects, SLSA uses a canonical source repository approach where users of open-source software need to only trust a handful of build platforms instead of trusting thousands of developers who have upload permissions to many packages. SLSA would simply reject an upload if it was not from a canonical source repository. And finally, for users of vendor-provided software, SLSA would require a vendor to be certified by a third-party auditor. This would make it easier for consumers to trust vendor software products. Also, provenance is the main aspect of using SLSA, it would provide traceability and verifiability of where the software originated. The goal is not just to prevent tampering but also to trace who was involved and where the exact origin of the software was. Since it facilitates transparency, companies can easily adapt to regulatory restrictions while ensuring that their business operations are sound. Hardening the Chain of Trust Hardening the Chain of Trust The core of SLSA is provenance , which documents the origin and security of artifacts. SLSA can pinpoint tampering or any other malicious behavior using cryptographic validation and automated attestation. Since it verifies that each step from creation to distribution of the software is indeed secure, the chain of trust is therefore solidified. provenance provenance The majority of well-secured development environments apply the principles of provenance including Google. One of the innovators of the SLSA framework at Google is Akash Mukherjee. He discusses how Chrome ensures verifiability in its builds in this article Chromium Chronicle #23 . The article focuses on the fundamentals of the framework and the four pillars that we see today in SLSA. Chromium Chronicle #23 Chromium Chronicle #23 Meeting Government Mandates With SLSA Meeting Government Mandates With SLSA Another advantage of SLSA compared to other security libraries is its ability to meet U.S. government regulations. The National Institute Of Standards and Technology aka NIST plays a crucial role in cybersecurity by creating frameworks, guidelines, etc. like the Secure Software Development Framework ( SSDF ). SLSA is able to help companies meet these requirements. SSDF SSDF Following this framework will also align companies with the Executive Order 14028 directive which has a higher standard for cybersecurity along the entire software supply chain. Also in highly regulated defense, healthcare, and finance industries, SLSA is reliable enough to be the security blueprint for organizations that can help generate evidence of compliance even when it touches critical cybersecurity concerns. The Future of SLSA as a Blueprint for Secure Software Development The Future of SLSA as a Blueprint for Secure Software Development But this journey does not stop here; it continues evolving the industry. With the flexibility in its structure, SLSA allows organizations to enhance their security step by step while preparing for future threats. As SBOMs become central to software development, experts focus on securing development practices in general, “SBOMs can help lift security health from the status quo, but frameworks like Supply-chain Levels for Software Artifacts (SLSA) and Secure Software Development Framework (SSDF) must be used to improve security as you’re building new packages.” , Akash Mukherjee, an early developer of the SLSA framework in his recent post on RSA conference. “SBOMs can help lift security health from the status quo, but frameworks like Supply-chain Levels for Software Artifacts (SLSA) and Secure Software Development Framework (SSDF) must be used to improve security as you’re building new packages.” recent post recent post By integrating the best practices from SLSA with development insights gleaned from resources such as Chromium Chronicle, an organization is well-equipped to design secure software. At its core, SLSA revolves around provenance, providing cryptographic assurances and compliance with regulatory requirements, making it an extremely important tool for securing the global software ecosystem.

How SLSA Can Secure Your Software Supply Chain

Exploring Product Insights: Interview with Vadim Glazkov, Founder of a Product Research Agency

Exploring Product Insights: entrevista amb Vadim Glazkov, fundador d'una agència d'investigació de productes

Explorando información sobre productos: entrevista con Vadim Glazkov, fundador de una agencia de investigación de productos

Esplorare le informazioni sui prodotti: intervista con Vadim Glazkov, fondatore di un'agenzia di ricerca sui prodotti

製品インサイトの探究: 製品調査会社創設者 Vadim Glazkov 氏へのインタビュー

제품 통찰력 탐색: 제품 연구 기관 설립자 Vadim Glazkov와의 인터뷰

Produktų įžvalgų tyrinėjimas: pokalbis su Vadimu Glazkovu, Produktų tyrimų agentūros įkūrėju

Истраживање увида у производе: Интервју са Вадимом Глазковим, оснивачем агенције за истраживање производа

Önüm düşünjelerini öwrenmek: Önüm gözleg gullugynyň esaslandyryjysy Wadim Glazkow bilen söhbetdeşlik

Exploring Product Insights: інтерв’ю із Вадимом Глазковим, засновником агентства Product Research

پروڈکٹ کی بصیرت کی تلاش: پروڈکٹ ریسرچ ایجنسی کے بانی، وادیم گلازکوف کے ساتھ انٹرویو

Mahsulot tushunchalarini o'rganish: Mahsulot tadqiqot agentligi asoschisi Vadim Glazkov bilan suhbat

Ukuphonononga ulwazi lweMveliso: Udliwano-ndlebe noVadim Glazkov, uMseki we-Arhente yoPhando lweMveliso

Google's Gemini AI: A Thought Experiment in Deception

Google ukan Gemini AI: ¿Mä amuyt’äw yant’äwix sallqjañataki?

La inteligencia artificial Gemini de Google: ¿un experimento mental sobre el engaño?

AI של Gemini של גוגל: ניסוי מחשבה בהונאה?

Google-ի Gemini AI. Մտքի փորձ խաբեության մեջ:

Google の Gemini AI: 欺瞞の思考実験?

구글의 제미니 AI: 기만에 대한 사고실험?

Google-ийн Gemini AI: Хууран мэхлэлтийн талаархи бодлын туршилт уу?

Gemini AI Google: Eksperimen Pemikiran dalam Penipuan?

د ګوګل جیمني AI: په فریب کې د فکر تجربه؟

Google Gemini AI: Tijaabo Fikirka ah ee Khiyaanada?

Gemini AI ya Google: Jaribio la Mawazo katika Udanganyifu?

Trí tuệ nhân tạo Gemini của Google: Một thí nghiệm tư duy về sự lừa dối?

It's Not A Motivation Problem That You Have (It's a Clarity Problem)

In the beginning, there was and would always be a startup founder — the Creator of the new business universe. Besides being tough, entrepreneurial beginnings can be very lonely endeavors. You need a team for your new startup to survive and thrive. These people joining the founder, usually the Cs (CTO, CFO, COO, etc.), are the startup's "core." They have to wear more than one hat and take all the heat until catching the right startup beat. Think of a startup as a tree that's growing and branching. New branches mean new people on board. The Book of Startup Genesis Let's get back quickly to the "core." At the very beginning, the startup creator (founder) is also the chooser. They choose who they're going to hire. At one point, someone else is going to do it for them. This is not a big deal when a startup is transitioning to a company. However, it can become the "thing" when a startup finds itself in deep corporate waters, usually during the IPO phase. The startup founders have a lot on their plate to micromanage, let alone worry about hiring or firing people. Now, I feel it'd be the right moment to address the elephant in this article. What qualifies me to tackle this topic in the first place? Well, it's not the fact that I've tried and failed, but what came as the aftermath of my personal startup tryouts. I have lost count of how many entrepreneurs and startup founders hired me as a freelance content writer or growth marketer since 2011. Some made it, and it's not my credit. Many failed, and it's not my fault. I couldn't help but notice a pattern. Do you, by any chance, remember how Tolstoy's "Anna Karenina" begins? It's this line: Happy families are all alike; every unhappy family is unhappy in its own way. When it comes to startups, it's the exact opposite: Unsuccessful startups are all alike; every successful startup is successful in its own way. Somewhere along the way, startup founders have it too much, and they're out of touch with both people and operations at the lower levels. It's always been like this, nothing personal it's just business - ignorance is (not) bliss or an excuse. These Are Just My Opinions - I Used To Be One of The Startup Minions I've never had problems working with and for startup founders directly, not even remotely (both literally and figuratively regarding the last word of this line). But, with COOs and other "minions," it's been a completely different story. It's not fair that startup founders are unaware by default, but again, that's just the way business things are. You have every right to criticize my (mis)use of the term "minions" in a startup concept. You can also label everything that you've read so far as hypothetical and abstract. So, let's bring some well-known real-life startup business examples. Shall we? Do you want to sell sugared water for the rest of your life? Or do you want to come with me and change the world? It's such a relief when you use a quote with no need to mention who said it because everybody knows it already. It's tricky to "minionize" a person who masterminded the legendary Apple "1984" commercial. How it started... I think of you just like Woz and Markkula. You're like one of the founders of the company. They founded the company, but you and I are founding the future. — Steve Jobs, to John Sculley, at the Macintosh introduction rehearsal in 1984 How it ended... Sculley said in 2015 that Jobs never forgave him and their friendship was never repaired. I don't know. I can only speculate. But, if the Pepsi guy had the audacity to axe the biggest apple on the Apple tree, I could only imagine what happened to the "low-hanging" fruits loyal to Steve Jobs. Here's another one. Justin Timberlake portrayed Sean Parker in "The Social Network." This guy sure thing knew how to throw memorable parties. Despite drawing, I mean doing some "lines," he and Zuckerberg stayed close (met regularly). "In 2017, during an interview with Axios, Parker expressed concerns about the role of Facebook in society, saying that it "exploit[s] a vulnerability in human psychology" as it creates a "social-validation feedback loop". Parker stated that he was "something of a conscientious objector" to using social media." So, my dear startup founders, do you know what your minions are doing while you aren't looking? In the beginning, there was and would always be a startup founder — the Creator of the new business universe. Besides being tough, entrepreneurial beginnings can be very lonely endeavors. You need a team for your new startup to survive and thrive. These people joining the founder, usually the Cs (CTO, CFO, COO, etc.), are the startup's "core." They have to wear more than one hat and take all the heat until catching the right startup beat. Think of a startup as a tree that's growing and branching. New branches mean new people on board. The Book of Startup Genesis Let's get back quickly to the "core." At the very beginning, the startup creator (founder) is also the chooser. They choose who they're going to hire. At one point, someone else is going to do it for them. This is not a big deal when a startup is transitioning to a company. However, it can become the "thing" when a startup finds itself in deep corporate waters, usually during the IPO phase. The startup founders have a lot on their plate to micromanage, let alone worry about hiring or firing people. Now, I feel it'd be the right moment to address the elephant in this article. What qualifies me to tackle this topic in the first place? Well, it's not the fact that I've tried and failed, but what came as the aftermath of my personal startup tryouts. I have lost count of how many entrepreneurs and startup founders hired me as a freelance content writer or growth marketer since 2011. Some made it, and it's not my credit. Many failed, and it's not my fault. I couldn't help but notice a pattern. Do you, by any chance, remember how Tolstoy's "Anna Karenina" begins? It's this line: Happy families are all alike; every unhappy family is unhappy in its own way. Happy families are all alike; every unhappy family is unhappy in its own way. When it comes to startups, it's the exact opposite: Unsuccessful startups are all alike; every successful startup is successful in its own way. Unsuccessful startups are all alike; every successful startup is successful in its own way. Somewhere along the way, startup founders have it too much, and they're out of touch with both people and operations at the lower levels. It's always been like this, nothing personal it's just business - ignorance is (not) bliss or an excuse. These Are Just My Opinions - I Used To Be One of The Startup Minions I've never had problems working with and for startup founders directly, not even remotely (both literally and figuratively regarding the last word of this line). But, with COOs and other "minions," it's been a completely different story. It's not fair that startup founders are unaware by default, but again, that's just the way business things are. You have every right to criticize my (mis)use of the term "minions" in a startup concept. You can also label everything that you've read so far as hypothetical and abstract. So, let's bring some well-known real-life startup business examples. Shall we? Do you want to sell sugared water for the rest of your life? Or do you want to come with me and change the world? Do you want to sell sugared water for the rest of your life? Or do you want to come with me and change the world? It's such a relief when you use a quote with no need to mention who said it because everybody knows it already. It's tricky to "minionize" a person who masterminded the legendary Apple "1984" commercial. How it started... How it started... I think of you just like Woz and Markkula. You're like one of the founders of the company. They founded the company, but you and I are founding the future. — Steve Jobs, to John Sculley, at the Macintosh introduction rehearsal in 1984 I think of you just like Woz and Markkula. You're like one of the founders of the company. They founded the company, but you and I are founding the future. — Steve Jobs, to John Sculley, at the Macintosh introduction rehearsal in 1984 How it ended... How it ended... Sculley said in 2015 that Jobs never forgave him and their friendship was never repaired. Sculley said in 2015 that Jobs never forgave him and their friendship was never repaired. I don't know. I can only speculate. But, if the Pepsi guy had the audacity to axe the biggest apple on the Apple tree, I could only imagine what happened to the "low-hanging" fruits loyal to Steve Jobs. Here's another one. Justin Timberlake portrayed Sean Parker in "The Social Network." This guy sure thing knew how to throw memorable parties. Despite drawing, I mean doing some "lines," he and Zuckerberg stayed close ( met regularly ). met regularly "In 2017, during an interview with Axios , Parker expressed concerns about the role of Facebook in society, saying that it "exploit[s] a vulnerability in human psychology" as it creates a "social-validation feedback loop". Parker stated that he was "something of a conscientious objector" to using social media." Axios So, my dear startup founders, do you know what your minions are doing while you aren't looking? So, my dear startup founders, do you know what your minions are doing while you aren't looking?

 Dear Startup Founders: Beware of Your Minions!

 প্রিয় স্টার্টআপ প্রতিষ্ঠাতা: আপনার মিনিয়নদের থেকে সাবধান!

 Vážení zakladatelé startupů: Dejte si pozor na své přisluhovače!

Estimados fundadores de startups: ¡Cuidado con sus secuaces!

 Estimados fundadores de startups: Coidado cos teus Minions!

प्रिय स्टार्टअप संस्थापकों: अपने गुर्गों से सावधान रहें!

 Հարգելի ստարտափի հիմնադիրներ. Զգուշացեք ձեր մինիոններից:

スタートアップの創業者の皆様へ: 手下にご注意ください!

Beste startup-oprichters: pas op voor uw minions!

 Nshuti bashinze gutangiza: Witondere Minions zawe!

 Vanodiwa Vatangi vekutanga: Chenjererai Maminions enyu!

 Kära startup-grundare: Se upp för dina minions!

 Шановні засновники стартапів: стережіться своїх прихильників!

Easy Ways to Learn a Language With ChatGPT

Stories, News & Mentions about HYPOTHETIC

Hypothetic’s Stories on HackerNoon