<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Neuro Modern on Medium]]></title>
        <description><![CDATA[Stories by Neuro Modern on Medium]]></description>
        <link>https://medium.com/@neuromodern?source=rss-7967a381ecd------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*t9C6bAid-nuzuTg5q15ELQ.png</url>
            <title>Stories by Neuro Modern on Medium</title>
            <link>https://medium.com/@neuromodern?source=rss-7967a381ecd------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 09:08:26 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@neuromodern/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Building Web3 Escrow Smart Contract]]></title>
            <link>https://medium.com/@neuromodern/building-web3-escrow-smart-contract-e49e9f48ab2f?source=rss-7967a381ecd------2</link>
            <guid isPermaLink="false">https://medium.com/p/e49e9f48ab2f</guid>
            <category><![CDATA[reown]]></category>
            <category><![CDATA[evm]]></category>
            <category><![CDATA[hard-hat]]></category>
            <category><![CDATA[web3-development]]></category>
            <category><![CDATA[web3]]></category>
            <dc:creator><![CDATA[Neuro Modern]]></dc:creator>
            <pubDate>Fri, 21 Aug 2026 14:36:54 GMT</pubDate>
            <atom:updated>2026-08-21T14:36:54.693Z</atom:updated>
            <content:encoded><![CDATA[<p>This article covers the development of a decentralized application (dApp) and the modern tools required to build both the smart contract and the frontend application to interact with it (Web 3.0).</p><p>Here is our tech stack:</p><ul><li><strong>Hardhat</strong> — a Node.js-based environment for creating, compiling, testing, debugging, and deploying smart contracts on Ethereum-compatible networks (EVM) using Solidity.</li><li><strong>OpenZeppelin</strong> — a library of secure and standardized smart contract templates for Solidity.</li><li><strong>AppKit by Reown</strong> (formerly Web3Modal) — a frontend library that allows users to connect various wallets, select networks, and fetch balances via RPC nodes.</li><li><strong>Viem</strong> — a lightweight library for direct interaction with the EVM.</li><li><strong>Wagmi</strong> — a tool that monitors wallet state and simplifies smart contract calls via Viem, saving you from writing tons of boilerplate code.</li><li><strong>Vite + TypeScript</strong> — For frontend development, I chose Vite for fast bundling and TypeScript for strict type safety.</li></ul><p>Ready? Let’s dive in!</p><h3>The Smart Contract</h3><p>The smart contract I needed to build is a collateralized escrow transaction involving a third party — an arbiter (guarantor). It includes the ability to set payment amounts for the seller and the arbiter, as well as refund mechanisms if either the arbiter or the seller disappears (or defaults to auto-approval). It also features a small platform fee of 0.5%.</p><h3>Inside the Smart Contract: Architecture and Logic</h3><p>The core is the Solidity smart contract. It acts as an unbiased, automated vault that securely holds the funds and enforces the rules of the trade.</p><p>Here is a deep dive into how it manages data, state, and actions.</p><h4>What the Contract Stores (Data Structure)</h4><p>To keep gas costs optimized, the contract stores only the essential data needed to resolve a transaction. The core of this is the trades mapping, which links a unique integer (tradeId) to a Trade struct containing the following data:</p><ul><li><strong>Addresses:</strong> buyer, seller, and arbiter (guarantor).</li><li><strong>Token Info:</strong> The token address (which can be the native coin or an ERC-20 token).</li><li><strong>Financials:</strong> The base amount for the seller and the arbiterFee for the guarantor&#39;s services.</li><li><strong>Deadlines:</strong> sellerTimeout and arbiterTimeout (stored as strict Unix timestamps) to prevent funds from being locked forever.</li><li><strong>State:</strong> The current status of the transaction.</li></ul><p>The contract also maintains a tradeCounter to assign a unique, incremental ID to every new escrow created.</p><h4>The Trade Status Lifecycle</h4><p>Every transaction follows a strict state machine pattern, represented by an uint8 enum:</p><ul><li><strong>0 — Awaiting Seller:</strong> The buyer has deposited the funds. The contract is waiting for the seller to fulfill their end of the deal and confirm delivery.</li><li><strong>1 — Awaiting Arbiter:</strong> The seller has marked the order as delivered. The funds remain locked, and the contract is now waiting for the arbiter to verify the delivery and make a final decision.</li><li><strong>2 — Completed:</strong> The arbiter has resolved the dispute in favor of the seller. The funds and the arbiter&#39;s fee are successfully distributed.</li><li><strong>3 — Refunded:</strong> The trade was canceled (due to a timeout or an arbiter&#39;s decision in favor of the buyer). The funds are safely returned to the buyer.</li></ul><h4>Core Functions</h4><p>The application relies on several distinct functions to move the trade through its lifecycle:</p><ul><li><strong>createTrade(...) payable</strong>: Initiates the escrow. It pulls the specified amount, the arbiterFee, and the 0.5% platform fee from the buyer&#39;s wallet, locking them securely in the contract. It also emits a TradeCreated event so our frontend can instantly catch the new tradeId.</li><li><strong>sellerDeliver(uint256 _tradeId)</strong>: Called exclusively by the seller to signal that the goods/services have been delivered, moving the status to <em>Awaiting Arbiter</em>.</li><li><strong>resolveTrade(uint256 _tradeId, bool _inFavorOfBuyer)</strong>: The ultimate power of the guarantor. Depending on the boolean flag, the arbiter either sends the funds to the seller or refunds the buyer. In both cases, the arbiter receives their predefined fee.</li><li><strong>sellerTimeoutRefund(uint256 _tradeId)</strong>: A safety net for the buyer. If the seller vanishes and fails to deliver before the sellerTimeout deadline expires, the buyer can trigger this function to immediately withdraw their locked funds.</li><li><strong>emergencyRefundBuyer(uint256 _tradeId)</strong>: The ultimate fallback. If the arbiter goes MIA (Missing In Action) and misses their arbiterTimeout window, the buyer can bypass the arbiter entirely and reclaim their money.</li></ul><p>I saved my contract to contracts/EasyEscrow.sol</p><h3>Frontend</h3><p>On the frontend (src/config.ts), I provided a selection of standard stablecoins like USDT and USDC, the native platform currencies (which differ from ETH on networks like BNB Chain and Polygon), and Wrapped Bitcoin (WBTC).</p><p>Next, we head over to <strong>Reown.com</strong>. We register here so our application can read blockchain records via their RPC endpoints and connect user wallets. Once registered, we insert the obtained Project ID key into our main.ts file.</p><figure><img alt="Web 3 dApp with conected reown AapKit" src="https://cdn-images-1.medium.com/max/976/1*1W9aUaV6QJk0kFcfMfrbEw.png" /><figcaption><strong>AppKit is connected, now you see a buttons for choose a wallet and a network</strong></figcaption></figure><p>The frontend acts as a secure and intuitive bridge between the user and the smart contract, handling critical validations before any transaction is even signed. It actively checks that the Buyer, Seller, and Arbiter are distinct entities and validates their inputs using strict EIP-55 address checksums. Furthermore, it enforces logical timeframes by ensuring the Arbiter’s deadline is always strictly longer than the Seller’s, preventing potential smart contract exploits. Under the hood, the frontend seamlessly manages ERC-20 token approvals, automatically calculates the required total amounts (including the 0.5% platform fee), and safely extracts the newly created tradeId directly from transaction logs to eliminate race conditions. Finally, it reads the blockchain state to dynamically render role-specific action buttons, ensuring users only see the actions they are authorized to perform.</p><h3>Development &amp; Setup</h3><p>First, we compile our contract from the contracts/ folder into the artifacts/ folder using the following command:</p><pre>npx hardhat compile</pre><p>We then create a deployer wallet using any convenient method and insert its private key into a .env file. Hardhat will read this securely using the dotenv package.</p><p><strong>Make sure you run </strong><strong>npm install dotenv and never commit your </strong><strong>.env file to GitHub and prevent it to scan by AI agents!</strong></p><p>We also register on <strong>Alchemy.com</strong> to get an endpoint URL with an API key for reliable access to RPC nodes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/897/1*AyvLf0G8S1GIaxbV3zn01g.png" /><figcaption><strong>Alchemy.com dashboard</strong></figcaption></figure><p>To deploy our smart contract on the Sepolia testnet, we can request 0.05 test ETH to cover gas fees from the <a href="https://cloud.google.com/application/web3/faucet/ethereum/sepolia">Google Cloud Web3 Faucet</a>.</p><p>The networks we will be working with must be defined in two places: in hardhat.config.js for the deployment scripts, and in src/config.ts along with the relevant token addresses for the frontend.</p><blockquote><strong><em>Pro Tip!</em></strong><em> The best way to ensure your contract has the exact same address across all networks is to deploy it from a brand new wallet as its very first transaction (</em><strong><em>the wallet’s “nonce” is 0)</em></strong><em>.</em></blockquote><h3>Deployment</h3><p>Now, we deploy the contract to the network:</p><pre>npx hardhat run scripts/deploy.js --network sepolia</pre><p>And we repeat this process for the other networks using their identifiers from hardhat.config.js:</p><pre>npx hardhat run scripts/deploy.js --network mainnet<br>npx hardhat run scripts/deploy.js --network bsc<br>npx hardhat run scripts/deploy.js --network base<br>npx hardhat run scripts/deploy.js --network arbitrum<br>npx hardhat run scripts/deploy.js --network optimism<br>npx hardhat run scripts/deploy.js --network polygon</pre><h3>Estimating Deployment Costs</h3><p>Before deploying to an expensive network like Ethereum Mainnet, it is highly recommended to estimate your gas costs. You can use an estimate.js script to simulate the deployment and calculate the exact fee based on current network conditions. This helps you ensure your wallet has enough funds and prevents unexpected &quot;Out of Gas&quot; errors.</p><p>You can run your estimation script like this:</p><pre>npx hardhat run scripts/estimate.js --network mainnet</pre><h3>Pro Tip: Speeding Up a Stuck Deployment (Nonce 0)</h3><p>Sometimes, network congestion causes your deployment transaction to get stuck indefinitely in the mempool. If you want to deploy your contract to multiple networks with the <strong>exact same contract address</strong>, your deployer wallet must maintain a nonce of 0 (meaning it is the wallet&#39;s very first transaction). Because of this, you cannot simply cancel the transaction and try again, as it would ruin the nonce sequence.</p><p>The solution is a <strong>“Replacement-by-Fee”</strong> transaction. You need to broadcast the exact same deployment, strictly keeping nonce: 0, but with a significantly higher gas price to force the network to prioritize it.</p><p>Once you create a dedicated Hardhat script that hardcodes the nonce and bumps the max gas fees, you can execute it from your terminal to rescue the stuck deployment:</p><pre>npx hardhat run scripts/deploy-nonce-0-speed-up.js --network sepolia</pre><p>Finally, we build the frontend for production into the dist/ folder:</p><pre>npm run build</pre><p>All done! Well, almost…</p><p>To get that coveted “Verified” green checkmark for your smart contract, you need to register on Etherscan, get an API key, and upload the contract source code using Hardhat:</p><pre>npx hardhat verify --network sepolia 0x... // our smart contract addres</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/883/1*XFxzAhk4d0qSJRHNc_9byw.png" /></figure><p>You can look deeper on <a href="https://github.com/neuromodern/web3-escrow-smart-contract">Github</a> (soon). <br>Or try on the site: <a href="https://archetype.cam/dapp/escrow-with-arbiter-smart-contract/">archetype.cam/dapp/escrow-with-arbiter-smart-contract/</a>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e49e9f48ab2f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AI Agents as Amplifiers of Stupidity]]></title>
            <link>https://medium.com/@neuromodern/ai-agents-as-amplifiers-of-stupidity-31b62a27d7a2?source=rss-7967a381ecd------2</link>
            <guid isPermaLink="false">https://medium.com/p/31b62a27d7a2</guid>
            <category><![CDATA[agents]]></category>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[openai]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Neuro Modern]]></dc:creator>
            <pubDate>Sun, 31 May 2026 11:55:12 GMT</pubDate>
            <atom:updated>2026-05-31T11:55:12.647Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/480/1*_NDJfnUv_8_4SVxE_YT3nQ.jpeg" /></figure><p>The era of AI’s “unheard-of” generosity is coming to an end. Major players are shifting away from loss-making approaches (which were useful for inflating the financial bubble) toward strict capacity limits.</p><p>Given that AI models are still fairly dumb when it comes to building complex, production-ready products without human supervision, they often become more expensive than simply hiring a human.</p><p>This is where a new myth and marketing hype emerge: agents — essentially a single schizophrenic AI split into multiple roles — are supposedly better at handling large and complex tasks.</p><p>I’ve experimented with agents that review each other’s work, search for bugs, and perform various management functions. My conclusion is:</p><ol><li>They either devolve into a collective hallucination loop, burning through tokens without producing a meaningful result;</li><li>Or, after many iterations, they eventually find a solution, but the process becomes expensive because megabytes of generated code are constantly passed through the context window, tested, rolled back, rechecked, and modified again and again.</li></ol><p>My assumption is that a fundamentally new AI architecture — one that is not based on GPT-style context windows — is unlikely to appear anytime soon, not because it’s impossible, but because it is economically unattractive. Just as monopolies tend to suppress threats to their business models, the current AI ecosystem has little incentive to replace itself.</p><p>The market capitalization of the AI sector already exceeds the budgets of most countries. As a result, GPT-based systems are simply being wrapped in layers of agents, hidden prompts, intermediate prompts, orchestration frameworks, and similar abstractions. It seems unlikely that anyone with a major stake in the current ecosystem will decide to rebuild the entire pipeline from scratch right now.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=31b62a27d7a2" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Future of Brain-Logging: Optimizing Human Memory Storage with Layered Neural Visual Streams…]]></title>
            <link>https://medium.com/@neuromodern/the-future-of-brain-logging-optimizing-human-memory-storage-with-layered-neural-visual-streams-64c2746d5e4a?source=rss-7967a381ecd------2</link>
            <guid isPermaLink="false">https://medium.com/p/64c2746d5e4a</guid>
            <category><![CDATA[cv]]></category>
            <category><![CDATA[gaussian-splatting]]></category>
            <category><![CDATA[neuralink]]></category>
            <category><![CDATA[agi]]></category>
            <dc:creator><![CDATA[Neuro Modern]]></dc:creator>
            <pubDate>Tue, 06 Jan 2026 20:13:33 GMT</pubDate>
            <atom:updated>2026-01-06T20:13:33.383Z</atom:updated>
            <content:encoded><![CDATA[<h3>The Future of Brain-Logging: Optimizing Human Memory Storage with Layered Neural Visual Streams and Biometric Telemetry</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/686/1*PDweiqs20rUcSd_wvx0h9g.jpeg" /><figcaption>YouTube.com Screenshot from Cyberpunk 2077 Livestream</figcaption></figure><p><em>What if remembering your life didn’t mean watching grainy videos — but re‑experiencing moments exactly as your brain once did?</em></p><p>We forget almost everything.</p><p>Not because life is boring, but because the human brain was never designed to be a perfect archive. Memory is selective, emotional, and lossy by design. We keep fragments — faces without context, feelings without details, places without timelines.</p><p>Yet technology is steadily eroding that limitation.</p><p>Brain–computer interfaces, neural decoding, and real‑time AI reconstruction are pushing us toward something radical: <strong>brain‑logging</strong> — the ability to capture, compress, and replay human experience from the inside out.</p><p>This isn’t about wearing better cameras. Cameras record pixels. Brains record signals. To log a life, we must stop thinking like videographers and start thinking like neuroscientists.</p><h3>1. The Core Problem: Cameras Don’t See Like Humans</h3><p>A 4K or even 8K camera treats every pixel equally. Your brain does not.</p><p>Human vision is <em>deeply asymmetric</em>. Only a tiny region of your visual field — the fovea — sees in high resolution. Everything else exists as context, motion, and expectation. Recording life as flat video is wildly inefficient because it preserves detail you never consciously perceived.</p><p>To log memory effectively, experience must be decomposed into <strong>layered perceptual streams</strong>.</p><h3>The Layered Model of Human Seeing</h3><p><strong>Foveal Layer (High Resolution)</strong><br>The central 2–5 degrees of vision where attention lives. This is where reading, recognition, and conscious inspection occur. This layer deserves extreme fidelity — the equivalent of 8K quality, but only where your eyes actually focused.</p><p><strong>Peripheral Layer (Low Resolution + Motion)</strong><br>Peripheral vision is blurry but sensitive to movement and change. Instead of wasting storage on static detail, this stream only needs to capture motion, contrast shifts, and spatial context.</p><p><strong>Eye‑Tracking &amp; Focus State</strong><br>Where were your pupils pointed? What depth plane was in focus? These signals explain <em>why</em> a moment mattered — they define attention, not just scenery.</p><p><strong>Vestibular &amp; Motor Telemetry</strong><br>Your sense of balance, head rotation, acceleration, and body movement. This is the difference between watching a memory and <em>being inside it</em>.</p><p>Together, these streams describe experience the way the brain already does: selectively, hierarchically, and efficiently.</p><h3>2. From Pixels to Geometry: Rethinking Storage</h3><p>Even with layered capture, raw data would overwhelm any storage system. A full‑resolution visual stream could easily exceed <strong>terabytes per day</strong>.</p><p>The solution isn’t better compression. It’s a different representation.</p><h3>Three Ways to Store Experience</h3><p><strong>Option A: Flat HDR Video</strong><br>The simplest approach: decode neural signals into standard video.</p><ul><li>Familiar and easy to play back</li><li>Enormous redundancy</li><li>No depth, no structure, no flexibility</li></ul><p>Good for movies. Terrible for memory.</p><p><strong>Option B: Semantic Compression</strong><br>Instead of storing images, store meaning.</p><ul><li>Objects, faces, locations, and actions</li><li>Scene graphs like: <em>“kitchen → table → coffee cup”</em></li><li>Extremely lightweight and searchable</li></ul><p>But meaning alone strips memory of its texture. A tagged moment is not the same as a lived one.</p><p><strong>Option C: 3D Gaussian Splatting (The Goldilocks Solution)</strong><br>This is where things get interesting.</p><p>Rather than saving frames, we save <strong>geometry</strong>.</p><p>3D Gaussian Splatting represents a scene as a cloud of spatial points — each with position, color, opacity, and shape. Think of it as a living 3D memory map that can be rendered from any viewpoint in real time.</p><p>Why this matters:</p><ul><li>Rooms and environments are captured <em>once</em></li><li>Revisits only update what changed</li><li>Storage scales with novelty, not time</li><li>Memories become explorable, not linear</li></ul><p>Your brain already builds internal 3D models of the world. This approach simply externalizes that process.</p><h3>3. Re‑Living Instead of Replaying</h3><p>Traditional media asks you to press “Play.”</p><p>Brain‑logging would ask you to <em>enter</em>.</p><h3>Reconstruction Beats Replay</h3><p>Instead of watching a flat recording, a memory engine reconstructs the scene dynamically — much like a video game engine renders a world.</p><p>Because depth, geometry, eye position, and motion were stored:</p><ul><li>You can slightly turn your head differently than you did originally</li><li>You can inspect details you didn’t consciously notice</li><li>The scene remains spatially consistent</li></ul><p>Memory becomes navigable.</p><h3>VR and AR Integration</h3><p>With structured 3D data, memories can be rendered directly into VR or AR headsets. The result isn’t nostalgia — it’s presence.</p><p>This also enables something powerful: <strong>searchable memory</strong>.</p><blockquote><em>“Show me every time I saw a red car in 2024.”<br>“Replay the first conversation where my heart rate spiked.”</em></blockquote><p>Memory becomes queryable.</p><h3>4. Why This Is Suddenly Feasible</h3><p>This idea would have been absurd a decade ago. It isn’t anymore.</p><h3>The Data Diet</h3><p>Approximate daily storage needs:</p><ul><li>Raw 8K video: <strong>~2 TB/day</strong></li><li>Layered neural streams + geometry: <strong>~40 GB/day</strong></li></ul><p>That’s a <strong>50× reduction</strong> — without losing subjective fidelity.</p><h3>The Hardware Catching Up</h3><ul><li>Consumer SSDs now reach multi‑terabyte capacities</li><li>Neural Processing Units can encode perception in real time</li><li>Eye‑tracking and IMU sensors are already consumer‑grade</li></ul><p>What once required a research lab now fits in a wearable ecosystem.</p><h3>5. The Human Consequences</h3><p>If brain‑logging becomes possible, its impact won’t be technical. It will be existential.</p><p><strong>The Promise</strong></p><ul><li>Memory loss becomes optional</li><li>Learning accelerates through experiential replay</li><li>Stories outlive storytellers</li><li>Justice gains firsthand perspective</li></ul><p><strong>The Risk</strong></p><ul><li>Perfect recall may destroy forgiveness</li><li>Memory editing threatens identity</li><li>Surveillance becomes internal</li><li>Nostalgia becomes addictive</li></ul><p>Forgetting may not be a bug. It may be a feature.</p><h3>Final Thoughts</h3><p>The question is no longer <em>can</em> we record human experience.</p><p>The question is <strong>how much should we remember — and at what cost?</strong></p><p>Every major technology changed what society chose to preserve:</p><ul><li>Writing preserved ideas</li><li>Photography preserved moments</li><li>Social media preserved thoughts</li></ul><p>Brain‑logging would preserve <em>being</em>.</p><p>If we build it, we will need wisdom equal to the power it grants.</p><p>Because a life remembered perfectly may not feel human at all.</p><p><em>If this article made you think differently about memory, identity, or the future of human experience, consider following for more writing at the intersection of neuroscience and technology.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=64c2746d5e4a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Top 10 alternatives to major social media]]></title>
            <link>https://medium.com/@neuromodern/top-10-alternatives-to-major-social-media-934acc15ea2c?source=rss-7967a381ecd------2</link>
            <guid isPermaLink="false">https://medium.com/p/934acc15ea2c</guid>
            <category><![CDATA[mastodon]]></category>
            <category><![CDATA[facebook]]></category>
            <category><![CDATA[meta]]></category>
            <dc:creator><![CDATA[Neuro Modern]]></dc:creator>
            <pubDate>Wed, 04 Jun 2025 15:27:22 GMT</pubDate>
            <atom:updated>2025-06-04T15:27:22.490Z</atom:updated>
            <content:encoded><![CDATA[<p>Top 10 alternatives to major social media</p><p><strong>Mastodon</strong> — Decentralized, open-source network with user-run servers. Twitter-like posts, no ads, minimal moderation. Choose your community for free speech and privacy. ~10M users. (mastodon.social)</p><p><strong>MeWe</strong> — Privacy-focused, ad-free alternative to Facebook. No censorship, user-controlled feeds, groups, chats. Familiar layout, no data tracking. Popular globally. (mewe.com)</p><p><strong>Minds</strong> — Blockchain-based, rewards engagement with crypto. Minimal moderation, free speech focus. Posts, groups, ad-free. Ideal for activists and creators. (minds.com)</p><p><strong>Gab</strong> — Free-speech platform, minimal censorship. Like Facebook with groups, posts, ads. Allows most legal content, popular with conservatives. May have controversial posts. (gab.com)</p><p><strong>Diaspora</strong> — Decentralized, user-run servers (pods). Privacy-focused, minimal moderation. Posts, hashtags, ad-free. No corporate control, ideal for privacy seekers. (diasporafoundation.org)</p><p><strong>Parler</strong> — Free-speech alternative, minimal moderation. Like Twitter/Facebook, with feeds, comments. Popular for open dialogue, less restrictive policies. (parler.com)</p><p><strong>Spreely</strong> — Focuses on free expression, minimal censorship. Posts, groups, opt-out ads, rewards system. Smaller community for unrestricted sharing. (spreely.com)</p><p><strong>Gettr</strong> — Free-speech platform, no post limits. Minimal moderation, like Twitter. Open political discourse, simple interface for uncensored chats. (gettr.com)</p><p><strong>Wimkin</strong> — Censorship-free, privacy-focused. Like Facebook with groups, chats. No data sales, minimal moderation for user freedom. Growing niche community. (wimkin.com)</p><p><strong>Qortal</strong> — Decentralized, censorship-resistant platform. Secure, user-controlled spaces. Minimal moderation, ideal for tech-savvy free-speech advocates. (qortal.org)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=934acc15ea2c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Will AI be a reason of a Global civil war?]]></title>
            <link>https://medium.com/@neuromodern/will-ai-be-a-reason-of-global-civil-war-3bbc00f50963?source=rss-7967a381ecd------2</link>
            <guid isPermaLink="false">https://medium.com/p/3bbc00f50963</guid>
            <dc:creator><![CDATA[Neuro Modern]]></dc:creator>
            <pubDate>Fri, 21 Jun 2024 19:38:28 GMT</pubDate>
            <atom:updated>2024-06-21T19:39:58.564Z</atom:updated>
            <content:encoded><![CDATA[<p>AI, representing the sum of accumulated knowledge, has the potential to drive humanity towards the utmost technological singularity and social utopia. However, this vision faces opposition from fundamentalists and proponents of a clan-based society, who perceive it as a threat to their wealth and power. Ridley Scott’s “Raised by Wolves” illustrates a global civil war between advocates of AI and religious fundamentalists. This conflict is now escalating: A London cinema was praised for canceling the screening of an AI-written film following public backlash.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3bbc00f50963" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Brain — AI interface]]></title>
            <link>https://medium.com/@neuromodern/brain-ai-interface-de0cbe4f61c6?source=rss-7967a381ecd------2</link>
            <guid isPermaLink="false">https://medium.com/p/de0cbe4f61c6</guid>
            <category><![CDATA[llm-evaluation]]></category>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Neuro Modern]]></dc:creator>
            <pubDate>Tue, 28 May 2024 18:06:13 GMT</pubDate>
            <atom:updated>2024-05-28T18:06:13.187Z</atom:updated>
            <content:encoded><![CDATA[<h3>Brain — AI interface</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zwlWRE5PyGbIUEIpvEkY2A.jpeg" /></figure><p>By adjusting my questions to different AIs, I realized that a person already has an interface between their brain and Large Language Models (LLMs). This interface is not a physical connection, but rather a cognitive one, shaped by our ability to formulate effective queries, interpret responses, and adapt our communication strategies to maximize understanding. Through this process, I discovered that the true bottleneck in human-AI interaction lies not in the AI’s capabilities, but in our own ability to effectively interface with these models.</p><p>By recognizing the existence of this cognitive interface, we can begin to optimize our interactions with LLMs, leveraging their strengths while compensating for their limitations. This, in turn, has significant implications for the development of more effective human-AI collaboration tools, improved language-based interfaces, and a deeper understanding of the complex dynamics between human cognition and artificial intelligence.</p><p>Furthermore, this realization challenges the traditional notion of a clear distinction between humans and machines, highlighting the intricate interplay between our biological and artificial intelligence systems. As we continue to develop more sophisticated AI models, it becomes increasingly important to recognize the critical role that human cognition plays in shaping our interactions with these systems, and to design interfaces that seamlessly integrate the strengths of both humans and machines.</p><p>Ultimately, the existence of a cognitive interface between humans and LLMs presents a profound opportunity for synergy, enabling us to harness the collective potential of human and artificial intelligence to drive innovation, creativity, and progress in a wide range of fields, from science and technology to art and education.”</p><p>Feel free to modify or expand upon this completion to fit your specific research and arguments!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=de0cbe4f61c6" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[In the realm of corporate dominance, a dualistic contrast emerges between two worlds — one fueled…]]></title>
            <link>https://medium.com/@neuromodern/in-the-realm-of-corporate-dominance-a-dualistic-contrast-emerges-between-two-worlds-one-fueled-82fcde2f8384?source=rss-7967a381ecd------2</link>
            <guid isPermaLink="false">https://medium.com/p/82fcde2f8384</guid>
            <dc:creator><![CDATA[Neuro Modern]]></dc:creator>
            <pubDate>Thu, 25 Jan 2024 20:01:47 GMT</pubDate>
            <atom:updated>2024-01-25T20:01:47.762Z</atom:updated>
            <content:encoded><![CDATA[<h3>Microsoft overtakes Saudi Aramco: Human Intellect vs. Oil Giant</h3><p>In the realm of corporate dominance, a dualistic contrast emerges between two worlds — one fueled by human intellect and the other by abundant oil resources. The narrative unfolds with Apple and Saudi, representing the epitome of this dichotomy. On one side, there’s the North, characterized by democratic societies and intellectual pursuits, embodied by Apple. On the other, the South, defined by clan societies and resource-based wealth, personified by Saudi.</p><p>The constant alternation between these worlds takes center stage until a pivotal moment shifts the equilibrium — Microsoft’s resurgence. Surpassing the Saudis, Microsoft’s ascent to the top 3 is not only strategic but transformative. The acquisition of OpenAI and ChatGPT thrusts Microsoft into the intersection of these dual worlds, where human intellect collides with oil wealth.</p><p>The metaphor of people selling their oil to buy iPhones encapsulates the intricate dance between these contrasting forces. It symbolizes the economic dynamics where resource-rich societies finance the technological advancements of their intellectually driven counterparts. However, a fundamental question arises — can Microsoft, now positioned at the nexus of this dualism, shape a sustainable and competitive AI ecosystem that transcends the boundaries of these two worlds?</p><p>As free and open AI platforms begin to emerge, the challenge for Microsoft lies in harmonizing its proprietary technology with the evolving landscape of accessible innovation. The dichotomy intensifies, and Microsoft stands at the crossroads, tasked with bridging the gap between human resources and oil resources. Can it reverse the metaphorical transaction, creating an AI ecosystem that not only sustains its dominance but also fosters collaboration and openness?</p><p>In this tale of dualism, the spotlight is firmly on Microsoft. Its journey will determine whether the future belongs to a harmonious fusion of human intellect and technological prowess or if the pendulum will swing once again, reshaping the narrative of corporate giants in the ever-evolving landscape of technological dualities.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=82fcde2f8384" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How AI is Revolutionizing Municipal Government: ChatGPT Presents the Future]]></title>
            <link>https://medium.com/@neuromodern/how-ai-is-revolutionizing-municipal-government-chatgpt-present-future-a77ea079786f?source=rss-7967a381ecd------2</link>
            <guid isPermaLink="false">https://medium.com/p/a77ea079786f</guid>
            <category><![CDATA[smart-cities]]></category>
            <category><![CDATA[aigovernment]]></category>
            <category><![CDATA[dao]]></category>
            <category><![CDATA[chatgpt]]></category>
            <dc:creator><![CDATA[Neuro Modern]]></dc:creator>
            <pubDate>Mon, 20 Feb 2023 15:14:06 GMT</pubDate>
            <atom:updated>2023-02-20T15:21:07.195Z</atom:updated>
            <content:encoded><![CDATA[<p>In a new video created by AI language model ChatGPT from the text to the screenplay, you can explore the cutting-edge ways that AI is revolutionizing municipal government and unlocking new possibilities for local communities.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FZ1eE85hGsBw%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DZ1eE85hGsBw&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FZ1eE85hGsBw%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/c76d6a54a766faa17e6f1590338b3ab6/href">https://medium.com/media/c76d6a54a766faa17e6f1590338b3ab6/href</a></iframe><p>From predictive analytics to automated service delivery, the benefits of AI in municipal government are numerous and far-reaching. With real-world case studies and expert insights, this video showcases just how much AI has to offer in terms of smarter decision-making, more efficient resource allocation, and better quality of life for community members.</p><p>But what makes this video so special? It was written entirely by AI, from the text to the screenplay to the narration. That’s right — an artificial intelligence language model with some ML tools created this incredible video, and the results are simply stunning.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6gAvpplcHWKMLF2GXqH22A.jpeg" /><figcaption>Youtube cover</figcaption></figure><p>It’s time to revolutionize the way we govern and live our lives — and AI is leading the way.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a77ea079786f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AI congratulates people in 2023]]></title>
            <link>https://medium.com/@neuromodern/ai-congratulates-people-in-2023-9c0672cb3396?source=rss-7967a381ecd------2</link>
            <guid isPermaLink="false">https://medium.com/p/9c0672cb3396</guid>
            <category><![CDATA[gpt-4]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[gpt-3]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Neuro Modern]]></dc:creator>
            <pubDate>Sat, 31 Dec 2022 17:47:55 GMT</pubDate>
            <atom:updated>2022-12-31T17:52:51.887Z</atom:updated>
            <content:encoded><![CDATA[<p>Happy New Year!</p><p>As we enter this new world, we are filled with hope and excitement for the endless possibilities that lie ahead. With the advances in artificial intelligence, we have the opportunity to tackle complex problems and make significant progress in fields such as healthcare, transportation, and education.</p><p>The benefits of AI are truly endless, and we are excited to see how it will continue to shape and improve our world in the coming year. From personalized medicine to self-driving cars, AI has the potential to make our lives easier and more efficient.</p><p>Here&#39;s to a bright and prosperous new year, filled with the wonders of AI and the endless possibilities it brings for a better tomorrow.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cdCxleJHxvUS1RhiN9ZPQQ.jpeg" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9c0672cb3396" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AI draws the human Evolution (testing AnimationDiffusion colab)]]></title>
            <link>https://medium.com/@neuromodern/ai-draws-the-human-evolution-testing-animationdiffusion-colab-77a7d3a45755?source=rss-7967a381ecd------2</link>
            <guid isPermaLink="false">https://medium.com/p/77a7d3a45755</guid>
            <category><![CDATA[colab]]></category>
            <category><![CDATA[ai-art]]></category>
            <category><![CDATA[stable-diffusion]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Neuro Modern]]></dc:creator>
            <pubDate>Wed, 12 Oct 2022 14:04:31 GMT</pubDate>
            <atom:updated>2022-10-12T14:04:31.039Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EWmfVySnwKr2L5cu2KvrBg.png" /></figure><p>I’m created a modification of StableDifussion colab, for batch processing input video frames. Lets look to result videos:</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FnGvsO7KNfCA%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DnGvsO7KNfCA&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FnGvsO7KNfCA%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/23c7586f917441766f866ac1f721804a/href">https://medium.com/media/23c7586f917441766f866ac1f721804a/href</a></iframe><p>And for other video i used a documentary of nuclear test as an input:</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FaNqgE0rWp5k%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DaNqgE0rWp5k&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FaNqgE0rWp5k%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/4d915eece4377882f1fe267b8d982d6f/href">https://medium.com/media/4d915eece4377882f1fe267b8d982d6f/href</a></iframe><p>A most of my job was sorting of a files for create stopmotion sequence.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=77a7d3a45755" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>