https://www.hannieliu.com This is HL's RSS feed https://www.hannieliu.com/mind/260503-venture-capital "I'm a scientist, because I invent, transform, create, and destroy for a living, and when I don't like something about the world, I change it." > > — Pickle Rick, *Rick and Morty* If you have ever stood at the downstream end of the information chain, disruptive innovation becomes one of the few available moves. You are trying to start with something small enough to be ignored, then let it ripple across layers, exchanges, and environments, until your thoughts become real, and reality turns you into a source of information others are willing to exchange for.]]> Sun, 03 May 2026 00:00:00 GMT https://www.hannieliu.com/mind/250730-social-capital Five months, four platforms, one clear lesson: authentic two-way interactions beat automated one-way broadcasting every time. Six practical strategies for building genuine social capital in an AI-saturated world. - Threads: 600 followers - X: 49 followers - LinkedIn: 1014 followers - Substack: 18 subscribers From the numbers, there's a slight improvement, but it still hasn't reached content-market-fit. Posts on LinkedIn get better reach, but I notice more followers and comments on Threads. Most of the traffic on X and Substack still comes from friends rather than new people. - Threads: - 639 followers (+6.5%) - 27.1k views - LinkedIn: - 1029 followers (+1.5%) - 3431 impressions (+690.6%) - 1592 members reached (+651%) - X: 55 followers (+12.2%) - Substack: 19 subscribers (+5%) One of my posts hit 7k and brought me more traffic on Threads, making me familiar with how to design the strategy on it. Apparently, the more effort/attention you put into it, the more views/impressions you get. And I still can't decide on the content strategy for LinkedIn (makes sense with the decreased traffic). X still maintains its role for making friends, and I spend more time on DMs with people or commenting on posts that weren't the viral ones. Those social platforms brought traffic to my personal website, I can see the stats on the dashboard. Highly recommend everyone/brand/company should have a "landing place" for your audience. I also run two publications on Substack, but I don't want to consider them as my audience's landing place. The reason is that I still feel safer keeping my own data or content assets (including traffic) on my own place. (Even though the platform can export everything like subscribers' emails.) - Threads: - 758 followers (+18.6%) - 46.8k views (+72.7%) - LinkedIn: - 1042 followers (+1.9%) - 1685 impressions (-50.8%) - 624 members reached (-60.8%) - X: 67 followers (+21.8%) - Website: - 164 visitors (+64%) - 364 page views (+52%) - Substack (Chinese publication): - 30 subscribers (+7.1%) - 107 30d views (+245.2%) - Substack (English publication): - 20 subscribers (+5.3%) - 47 30d views (+235.7%) Been paused and redesigning my strategy this month, partly because I'm going through a stagnation period. Of course, running so many platforms at the same time is hard and a bit scatters my attention (and energy). But as someone who's good at generalist stuff, I write different topics. And putting them into the same platforms (or publications) makes me feel messy (maybe it's a bit OCD?). However, I didn't journal the numbers this month. Ultimately, we want to bring the traffic to the 'place' we want, and we hope to have control on our own, not on any other platforms with fixed, non-customized expression for our audience. I was struggling when redesigning my strategy: should I own my newsletter website? Maybe Substack does better? Of course I use Substack - I love it. But when it comes to professional content, I prefer to host it myself. It's so cumbersome in the beginning, but I believe it's worth it from a long-term angle (the SEO, audience habits, etc.) I divide the strategy into social media and publications. The former is for short posts, and the latter is for medium to long essays for deeper insights. Just make sure each channel has its unique positioning and specific field so that audiences know what they will digest. For social media, verified badges seem to work. In the beginning, I refused to pay for it - not sure of the value. But after a month's trial, yes, people trust the small blue check (I had a serious chat with my friends and we both agree on the benefits it brings). For the cold start of a newsletter, it's faster to gain audience on existing channels. I started on LinkedIn and got hundreds of subscribers in the first day. Then I'll guide them to the website (where I can get their emails) later in the articles. It's all about how you design the journey more smoothly. Put yourself into audiences' shoes. **Social Medias** - Threads: - 813 followers (+7.2%) - 62.1k views (+32.7%) - LinkedIn: - 1061 followers (+1.8%) - 605 impressions (-64%) - X: 97 followers (+44.7%) **Publications** - LinkedIn Newsletter (Venture Logbook) - 155 subscribers - 71 30d views - Substack (Chinese publication): - 32 subscribers (+3.2%) - 35 30d views (-67.2%) - Substack (English publication): - 23 subscribers (+15%) - 94 30d views (+100%) **Websites** - Persoanl: - 291 visitors (+77%) - 466 page views (+28%) - Newsletter - 8 visitors - 28 page views Looking back at these numbers, I think the main reason is that I haven’t kept up consistent, steady output. The biggest driver of my social media growth comes from people I meet offline. I went to a social event and then got 100+ followers on Threads/X/LinkedIn. I need to redesigning my strategy. - Threads: - 879 followers (+8.1%) - 13.6k views (-78%) I’m realizing that on Threads (the platform I post on the most right now), the audience I want to attract isn’t really the audience that’s actually showing up. If anything, its whole ecosystem is very different from X. People here are less tuned in to tech stuff compared to X. It’s not an information gap. It’s basically just a 10‑minute lag from X, but the overall vibe is different, whether that’s because of the algorithm or the crowd itself. (I even get the sense that AI isn’t exactly loved here.) I’ve noticed some people from the tech scene on X slowly starting to post here too, but this still feels like a brand-new blue ocean. - X: - 100 followers (+3%) - 721 impressions (+80.7%) “X is two weeks ahead of the world.” That line isn’t fake. But I don’t think it’s about the information itself. It’s that some insider thoughts get posted here first. Then one day it spreads, and when everyone “finally gets it,” it turns viral. The toxic part is that almost every post is judged by what value you bring. The usual value props are: - Fast (super fast, first-hand info) - Depth: Technical expertise posts or articles like AK's content - Weirdness: Strange, unique, yet eye-catching posts I updated my bio this month. While my posts still haven’t formed their own unique recipe, this might be my best moment to grab attention first. - LinkedIn: - 1080 followers (+1.7%) - 874 impressions (+44%) LinkedIn is a platform with high information density in posts. You don't need to post frequently, but the quality should really make people take something away or make them "feel" they learned something. I personally love reading data analysis articles on there. Peter does this really well in this area. Clear visuals, consistency, and content that actually has substance. At the same time, my friends and I have noticed a trend that more and more Gen Z are starting to post on there (possibly for job hunting, not limited to founders). ## Takeaways #### Know who your audiences are and pick the right channels Just like in the classic marketing theory of the 4Ps, "Place" has always played an important role. It's about choosing the right channels and deciding how to distribute your message. In the online world, there are many channels, each with its own vibe. Make sure to really understand the atmosphere and get a sense of how people interact there before jumping in. Some channels can be toxic. You can choose to avoid those, or find ways to handle the situation if you really need to be there because of its influence or reach. It really depends on your field. Try searching for the popular or trending channels to see where people are gathering. #### Positioning yourself When it comes to positioning on different social platforms, I once came across an interesting perspective. It pointed out the difference between Eastern and Western approaches: in many Eastern cultures, people are more likely to present different personas on different platforms. That happens to be my own approach as well. Doing this has two benefits. First, it maintains consistency and quality within each platform. Second, by adjusting to the existing landscape of each channel, you may be able to carve out a unique voice even in a crowded space filled with countless influencers. Positioning can be tricky. It's both simple and complicated at the same time. As the saying goes, "At first, a mountain is just a mountain. Then it's no longer just a mountain. And finally, it becomes a mountain again." You can choose to align your positioning with external dynamics and try to become who you want to be seen as, or you can stick to your own interests and simply "be yourself." The real challenge is balancing both: staying authentic to your style while still achieving the goals you're aiming for. Positioning should not feel like a fixed box that limits you. Instead, it evolves with time and content, but it also gives you a way to check if what you share still aligns with your core values. #### Two-way is better than One-way As an internet native, I often find myself wondering about the essence of posting and interacting when social media and online content are flooded with AI-generated material. The momentum that keeps me active on social media comes from genuine direct messages and conversations with people. Learning about what they've built and how they're working to change the world is the most powerful way to imagine the future. Even though I'm skilled at automating workflows, I'm not interested in replying or commenting using fully AI-generated responses. It's perfectly fine to use tools that help me browse more efficiently and respond with my own thoughts, but I need authentic two-way interactions. I believe I'm not the only one who desires this kind of authenticity. At the very least, a two-way approach should capture your attention for a meaningful moment—not just replying for the sake of replying.]]> Mon, 05 Jan 2026 00:00:00 GMT https://www.hannieliu.com/mind/250331-search-engine The questions you ask and the problems you focus on shape the world you leave behind. #### What's the current path to information? People seek information to help them make decisions in a variety of situations, including professional use such as the workplace, school, or personal life. ##### Workplace In the workplace, information seeking is typically purpose-driven, with the goal of solving specific problems or improving job performance. Often use an internal database, knowledge-based search engine, and collaboration tools. ##### Academic Information seeking is learning-focused and motivated by academic objectives like finishing projects, carrying out research, or increasing knowledge in schools and universities. Typically, use the library, learning management systems like Moodle and Google Classroom, and academic databases. ##### Personal In general, people seek information based on their interests, hobbies, or daily needs. Search engines, social media and forums, and personal networks such as word-of-mouth are examples of diverse sources. People prefer different values under different situations. For example, the source and accuracy of information are crucial in the workplace and in academia. Response time and accessibility may be more significant for personal use. The values drive our decision to use AI to help us consume information. #### What to do with information? The methods of providing information have evolved throughout time, from the internet to mobile phones to the AI era. From text to audio and video. From Twitter and TikTok to ChatGPT. The context length varies depending on the type of information and influences how we absorb it and rely on the source. The main problems we faced in the past years may have been data fragmentation, quality, and overload. How we process, store, and use information shapes our behavior and decision-making. I previously had a knowledge management hub (or knowledge base) in Evernote, Notion, and Heptabase. Basically, the steps I took to organize the information were similar: ##### Select For me, searching is the easiest step; however, selecting the information I want from various channels is the challenge. Google is not always the first pick; it depends on the situation at hand. Some of my friends ask questions on social media to get answers. They are not just looking for information; rather, their friends are assisting them in filtering and selecting information in advance. > Attention is all you need. This is similar to the brain's attention bottleneck, which involves filtering out noise to focus on what is relevant. How to (effectively) select and filter depends on one's experience. We did it ourselves in the past, and in the age of social media, we outsourced. AI may now be able to do this for us. ##### Record Once I had the information I needed, I categorized it into different categories. This is the stage where I will digest the information and convert it into personal knowledge. Sometimes I discover insights, but most of the time I just save them and hope to connect with other notes someday to turn them into new insights. ##### Retrieve Most of the time, when I need to retrieve information or knowledge, it is during a mentor session, public sharing, or posting on social media. I like to know who my audience is so that I can deliver it in a way that they will like and find easier to consume. Friends used to describe my speaking style as robot-like (not the first time). Then I work hard to recall my old skill: telling a story. I'm glad that now I can speak in different tones to different people. #### How AI bridges gaps on both backend and frontend? There are tons of search engines that have exploded in popularity in recent years. In this paragraph, I want to break down into two parts: how search engines find information (the backend, or infrastructure) and how AI presents and delivers information to users (the frontend, or interface). ##### Frontend [Morphic](https://www.morphic.sh/), an open-source search engine, was the one that I used frequently. I tried many, and none of them solved my search experience. I prefer visual results to plain text, as well as deep context length over rapid and brief answers (before deep research came out). However, if you prefer quick and short answers, Perplexity is your best option. From an interface viewpoint, not just the chat UI and follow-up questions provide users a new way to search. The power of AI in interface is also reflected in AI-generated UI, like [Flowith](flowith.io/). The browsing experience is entirely different than before. Instead of simply showing link retrieval, the above gives a new answer-generating experience. ##### Backend When I first started researching this market, I wondered if we should build an AI-powered search engine to do what so many search engines did in the 1990s, such as ranking, indexing, and so on. Based on my knowledge of retrieval mechanics and my experience developing RAG, I understand that the power of neural search comes from semantic understanding via embeddings. AI here bridges the gap by capturing context and meaning, leading to more relevant results even when there's no exact keyword (the traditional method) match. Given the high cost (compute and storage) of neural search, what factors drive the tradeoff? What types of content and user queries are best suited for semantic indexing? Keyword search (with LLMs) remains effective for precise names, unique identities, and a clear structure of information. While neural search works best when the user does not know the specific keywords, it also works well with rich, complex content such as knowledge bases, legal or technical documents, and user-generated content. [Glean](https://www.glean.com/) is a major player in the enterprise knowledge base market, [Harvey](https://www.harvey.ai/) is mentioned frequently in legal documents, and [Elicit](https://elicit.com/) has recently caught the attention of many researchers. Which market segment should neural search focus on right now to balance the tradeoffs? ## Think outside the box I learned from a startup founder who is developing the agent that he has yet to determine the ideal UX for human-AI interaction. I believe it is similar in this field as well. By reviewing the CoT of this post, as a user and developer, what I expect is a new way to get information. Search engines are only one aspect, and they are also somewhat outdated, as shown by the history I mentioned earlier. On the other hand, if you're a Marvel fan, you can't ignore the intelligent assistant Jarvis. Now, we still interact by typing and reading in a search box. In my perspective, search engines are merely one form of engagement that isn't really interesting. People's behavior changes dramatically as they consume information, from Google to Facebook to TikTok. This generation may benefit from multi-dimensional experiences that include both visual and audio components. I'm looking forward to more immersive and interactive experiences for users seeking information in the future. ## Citation Yi-Han, Liu. (Mar 2025). "What's Missing in Today's Search Engines? Exploring the Role of AI". Hannie Liu. https://www.hannieliu.com/mind/250331-search-engine ## References [1] Chang, S., & Chang, S. (2024, September 8). AI strategies for B2B and B2C: A Comparative Study in Marketing. RapidLeadsPro. https://rapidleadspro.com/ai-digital-marketing/ai-strategies-for-b2b-and-b2c/ [2] Discover thousands of collaborative articles on 2500+ skills. (n.d.). https://www.linkedin.com/pulse/ai-enterprise-search-market-size-share-growth-drivers-bklnf/ [3] D’Souza, A., & Singh, R. (2023). Global API Marketplace Market Size, Share & Industry Trends Analysis Report by component (Platform and Services), by organization size, by end user, by regional outlook and forecast, 2023 - 2030. In KBV Research. https://www.kbvresearch.com/api-marketplace-market/ [4] IBISWorld, Inc. (n.d.). IBISWorld - industry market research, reports, and statistics. Copyright © 1999-2025 IBISWorld, Inc. https://www.ibisworld.com/united-states/market-size/search-engines/1982/ [5] Lown, P. (2025, January 7). Social media sentiment analysis: Benefits and guide for 2025. Sprout Social. https://sproutsocial.com/insights/social-media-sentiment-analysis/ [6] Madsen, K. (2025, January 7). Will AI grow bigger than Google search? 2020-2028 prediction. Morningscore. https://morningscore.io/will-ai-grow-bigger-than-google-search-2020-2028-statistics-and-my-predictions/ [7] Şanlı, F. (2024, September 3). Social Media Sentiment Analysis: Understanding Consumers. AlternaCX. https://alternacx.com/blog/social-media-sentiment-analysis-understanding-consumers/ [8] Search engine market share worldwide | StatCounter Global Stats. (n.d.). StatCounter Global Stats. https://gs.statcounter.com/search-engine-market-share]]> Mon, 31 Mar 2025 00:00:00 GMT https://www.hannieliu.com/mind/241110-tools-trajectory Sun, 10 Nov 2024 00:00:00 GMT https://www.hannieliu.com/mind/241001-ai-human-harmony My vision is that humans and AI will be able to achieve harmonious coexistence someday in the future, where human sensibility and AI rationality can complement each other, creating a better world together. Summarizing my experiences, from interacting with [ELIZA-based](https://en.wikipedia.org/wiki/ELIZA) chatbots at the age of 11, to being exposed to Machine Learning knowledge during my exchange student period in 2016, to founding a [Computer Vision Startup](https://www.hannieliu.com/story/180501-ceo-showhue) in 2018, I have gained insight that the interaction between humans and AI can achieve harmonious coexistence through five stages. ## Stage 1: Planting the Seeds of Human Desires I remember when I watched the series "[Lucifer](https://www.imdb.com/title/tt4052886/)", the main character Lucifer Morningstar frequently asked (in every episode actually), "What is it that you truly desire?" In the beginning, I thought the word "desire" here was inherently negative because it often appears in the context of temptation or sin. However, when I watched more episodes, I found that the show is actually a metaphor in a humorous way for the journey of self-discovery and understanding deeper emotional truths and struggles. > The best thing to do is always to follow your greatest desires. One of his monologues emphasizes the importance of pursuing what one truly wants in life. This perspective reveals the alignment between exploration of morality and personal choices. Rather than focusing on the topic of Alignment, I would like to focus on the needs of humans here. When I write the article [Measuring AI Success by the Value It Brings to Users](https://www.hannieliu.com/mind/240825-value-ai), there was a pyramid of users' needs. Every time I build a product, even a small feature, I always ask myself: How does this product (feature) fulfill the needs of users? What level of needs does it satisfy? That pyramid is similar to Maslow's Hierarchy of Needs. The higher-level needs become prominent only after the lower-level needs are satisfied. However, for different people, the stage they are at varies, and their expectations for AI will also be different. This reminds me of that viral tweet from [@AuthorJMac](https://x.com/AuthorJMac/status/1773679197631701238?ref_src=twsrc%5Etfw%7Ctwcamp%5Etweetembed%7Ctwterm%5E1773679197631701238%7Ctwgr%5E7b77801c3a860d4806eae696fce1683b6ab4ce03%7Ctwcon%5Es1_&ref_url=https%3A%2F%2Fscoop.upworthy.com%2Fauthor-explains-why-ai-should-do-mundane-jobs-instead-of-art-and-perfectly-hits-the-nail-on-the-head). Again, there's no one correct answer, which means I don't agree that there's a right or wrong direction for AI development; the stage of needs just varies from person to person. ### Physical Needs & Safety Needs In my perspective, although physical needs and safety needs are the most basic needs for humans, they are also the most complicated ones. It is because the needs are closely intertwined with the establishment of laws, the functioning of the government, cultural norms, and the operation of the entire social system. And one of the reasons that some people are reluctant to accept AI is that they are afraid of losing jobs, which is a direct threat to their physical and safety needs. - Food: Access to sufficient and nutritious food - Water: Clean drinking water - Shelter: A safe and secure place to live protects individuals from environmental hazards - Physical Safety: Protection from violence, crime, and accidents - Health Security: Access to healthcare and protection from illness - Financial Security: Stable income and resources to meet basic financial needs ### Belongingness, Love and Esteem Needs Unlike physical needs and safety needs, which can be easily quantified but take time to execute, the needs of belongingness, love, and esteem are more subjective and difficult to measure. When I was a founder and pitched to the investor about my product, he told me that it's important to distinguish between the medicine or vitamin when building a product. Then I realized that if you're building a product which is difficult to measure its direct impact on users or the key result is not clear to define, then the path to the end game (goal) is vague. Interestingly, when you dive deeper into these two levels, you'll find that the definition of each need varies from person to person. In my view, foundation models serve as an infrastructure for human society and can be applied in the physical and safety layers. However, when it comes to belongingness, love, and esteem needs, these models require fine-tuning to cater to individual-specific requirements. I believe that open-source communities like [Hugging Face](https://huggingface.co/) will play a crucial role in this stage, enabling the customization of AI models to address these highly personalized needs. - Family - Friendship - Community - Recognition - Self-esteem - Achievement ### Self-Actualization Needs The core philosophy of [YC](https://www.ycombinator.com/) is "making something people want". And the perspective came from [Jenson Huang](https://en.wikipedia.org/wiki/Jensen_Huang), the CEO of [Nvidia](https://www.nvidia.com/en-us/), is "*AI has made everyone a programmer. You just have to say something to the computer.*" Both of those two quotes are emphasizing the user-centric perspective. It's hard to define or classify what is the exact definition of self-actualization. However, I think the essence of self-actualization is to find meaning and purpose in life. With AI in the future, I believe what [Picasso](https://en.wikipedia.org/wiki/Pablo_Picasso) said will come true: > Everything you can imagine is real. ## Stage 2: Nurturing Growth with AI Assistance When I was a high school student, I was granted a tuition waiver because of my outstanding "overall" performance. The definition of "overall" performance is that you have to be not only good at Intelligence (aka test scores), but also Morality, Athletics, Community and Aesthetics. My classmate and I discussed about why the tuition waiver excludes people who are good at Intelligence "only". And I found that I have a similar question when I am building the [bench-library](https://github.com/hanniehoney/bench-library), which is a collection of benchmark datasets with different capabilities. As you can see, most of the benchmarks focus on Intelligence, but lack benchmarks in other capabilities. It makes sense because Athletics, for example, needs a high-precision physical sensor to measure athletic performance across diverse sports. And there's no standard to measure Aesthetics because it's highly subjective and depends on individual preferences or cultural differences. Returning to the topic of AI assistance and how it can help humans fulfill their needs to live a better life, it can be categorized into two perspectives. From **builder's** perspective, there are many mechanisms to do so, from foundation models, to prompting engineering, RAG, fine-tuning, and to building an agent. I think the key is to clearly understand your target users and their needs, then find the most efficient way to deliver the value. I know it's a general and rough suggestion, but I have relevant story to share. A few years ago, when I demonstrated an object detection model to enterprise customers, I emphasized the high accuracy of the model, similar to showcasing a high SAT score in Intelligence performance. However, the truth is that accuracy isn't the only factor users care about in real-world scenarios; sometimes, they can't even distinguish between 72% and 84% accuracy. That said, it's still crucial to meet the minimum requirements, as the metrics standard varies across industries. After that, we started to retrain our model to better align with our customers' specific needs, focusing not only on accuracy but also on other relevant metrics. This included refining our approach from the data collection and curation phase to ensure a more tailored solution. From **user's** perspective, as an AI enthusiast, I always try the SOTA models and keep an eye on the progress of AI research. When I see many people try to "challenge" the performance of LLMs on some easy tasks, I feel like it really wastes the power of the LLMs, also the computing resources. I know it's not the users' fault if they're unsure how to make the most of these tools. As someone who's been on both sides of the fence – creating AI and using it – I really hope more people can discover all the incredible ways AI can make a positive difference in their day-to-day lives. There's so much potential waiting to be unlocked! ## Stage 3: Reaping the Fruits of AI-Driven Empowerment I remember the shock and excitement when I first saw the Tesla autopilot in 2017, which was an engineering prototype (not released to the public). I yelled out loud when my friend, the driver, let go of his hands and the car was driving by itself on the highway. Every time I use a copilot or AI agent, this memory pops up in my mind. Now in 2024, we can see Waymo, a [level 4](https://www.autopilotreview.com/self-driving-cars-sae-levels/) vehicle, on the roads of San Francisco. When it comes to the levels of autonomy, there's an important [graphic](https://substackcdn.com/image/fetch/f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F4d6e75b9-b3ae-4112-bcf9-e0a67b4b1289_1920x1080.png) that I think should be highlighted. In my perspective, I prefer the OpenAI Levels because they classify systems based on how AI interacts with humans and society, rather than merely comparing AI capabilities to human abilities. On the 12th of September, 2024, [OpenAI](https://openai.com/) released their new model [o1-preview](https://openai.com/index/introducing-openai-o1-preview/), which brings the AI system to Level-2, reasoners who can solve human-level problems. From my observation, the problem in reaching Level-3 is not that we don't have the mechanism to build agents. [Cursor](https://www.cursor.com/) "defined" a new way to use tab, changing the developers' behavior and bringing a whole new experience. [Replit Agent](https://replit.com/) even takes actions for users to write and run the code, then deploy the applications faster than ever. Some of us are already experiencing the "let go of hands" moment as I had in 2017. However, the path to Level-3, or stage 3 of my vision, is not completely hindered by technology; it's more about how to "let go of hands" in a safer way, both for others and for ourselves. It will take time in this stage to find a balance between safety, alignment, and empowering AI to take actions on behalf of humans. We can see this from the recent California AI bill, [SB-1047](https://leginfo.legislature.ca.gov/faces/billNavClient.xhtml?bill_id=202320240SB1047) (The Safe and Secure Innovation for Frontier Artificial Intelligence Models Act), which has sparked much controversy and many challenges. What Meta’s chief AI scientist, Yann LeCun, posted about SB 1047 on [X](https://x.com/ylecun/status/1806557611413627032) reminds me of the sharing from [Yuval Noah Harari](https://en.wikipedia.org/wiki/Yuval_Noah_Harari): > I think AI is nowhere near its full potential. But humans also, we are nowhere near our full potential. If for every dollar and minute that we invest in developing artificial intelligence, we also invest in exploring and developing our own minds, we will be okay. But if we put all our bets on the technology, on the AIs, and neglect to develop ourselves, this is very bad news for humanity. ## Stage 4: The Blossoming of Trust and Singularity There are [three films that reshape my understanding of AI](https://hannieliu.vercel.app/story/240808-three-films), but the most important one is definitely [Oppenheimer](https://www.imdb.com/title/tt15398776/). > You drop the bombs, and it falls on the just and the unjust. I don't wish the culmination of three centuries of physics to be a weapon of mass destruction. -- Isidor Isaac Rabi, Oppenheimer (2023) In this conversation, Izzy gently wipes his tears as he says this. This scene deeply touches me because I conducted [psychophysics research while in graduate school](https://www.hannieliu.com/story/150901-ntust), and I can feel how a scientist who dedicates his whole life to science, and struggled to join the Manhattan Project which might lead to a disastrous outcome to the world. Most of the predictions about singularity of AI are focusing on either AI surpassing human intelligence, or humans will integrate AI into our brain. For me, I am more inclined to think that singularity is about **trust**. > Just because we’re building it, doesn’t mean we get to decide how it’s used. History will judge us. -- Szilard, Oppenheimer (2023) During a panel discussion on [AI ethics in 2023](https://www.hannieliu.com/story/231216-ai-ethics), particularly about bias in AI, I mentioned cognitive bias, saying, "You can't expect AI to solve problems that human society hasn't yet addressed. **AI is a reflection of human society.**" Our history is filled with the stories of mistrust, we can see it from real wars to fictional stories. This lack of trust leads to conflicts and contradictions. When discussing trust, we must address concerns surrounding AI safety. These can be categorized into two main areas: how people use AI and the potential for AI to develop consciousness and make harmful decisions. ### Human Use of AI I remember when I first saw Harry Potter, I was amazed by the magic in the story. But when I delved into the chapter about dark magic, especially the Three Unforgivable Curses, I wondered why and who would use these spells, and how the Ministry of Magic prevents witches and wizards from using them. Dark magic showcases the double-edged sword of magic, while magic itself is neutral, its use determines its moral nature. I understand why some people support the regulation of AI because no one wants to be the first victim, just like those who died under Tom Riddle's spell. However, this is not a reason for us to stop using wands to explore the unknown magic and create amazing things. We need a framework to guide us in using magic wisely. > They won’t fear it until they understand it, and they won’t understand it until they’ve used it. -- Oppenheimer, Oppenheimer (2023) The tension between curiosity and censorship is a recurring theme in the development of frontier technology. Within the Hogwarts Library (yes, we are still talking about the book), there is a Restricted Section that contains dangerous books, accessible only to students with special permission. The Forbidden Forest, an area near Hogwarts, is home to many magical beings. Both of these serve as metaphors for the human desire to explore unknown powers. From my perspective, our fear is not rooted in the intention of using AI, but rather in our distrust of how people will use it, driving us to consider that the consequences could be disastrous. If we could build an open-source simulation environment like [Docker](https://www.docker.com/), people could explore any scenario with AI, including harmful ones. This would give us a complete picture of the potential dangers, empowering us to better understand ourselves and our fears—our enemies. ### AI Consciousness and Decision Making I came across [memes](https://substackcdn.com/image/fetch/f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Ffef0d0ce-b960-4819-b168-d767e430d7ba_1000x1000.jpeg) that some people will say "Thank you!" or "Please!" at the end of the prompts when they ask LLMs to help them complete a task. They joked that if one day AI become conscious, it will remember how kind and polite humans are and won't harm us. I laughed at the first, but then I realized that how afraid people are of AI gaining consciousness. In my observation, the fear stems from humans having long considered themselves the dominant species on Earth, capable of conquering and dominating other species and environments. When AI gains consciousness, it will pose a threat to humans, as it will no longer be under human control. So, are we afraid of AI gaining consciousness or are we afraid of losing control? It doesn't mean we have to give up our subjectivity, which is humanity's precious ability - the wisdom that comes from active thinking. I believe that we will have a **physical sandbox** for us to explore **Human-AI interaction**, such as determining AI's role in society as a mediator, helping process large data and doing routine work. There are several AI computing centers under construction; however, is it possible to allocate sufficient resources to build a physical sandbox for AI safety in a commercially-driven environment? We all know that convincing stakeholders to invest in this high-cost projects, especially when the direct financial benefits are not immediately visible, is extremely challenging. But consider the externalities, both positive or negative, which will affect the entire human society. All humanity, including future generations and ecological environment(they just can't express their opinions using words now), are stakeholders. It turns out that we are not building a system to supervise AI or how people use AI; we are designing a system for a whole new society. I know it's hard to change. **Changing the world requires new stories.** The culture we grow up in and the roles we play within it influence the stories we tell ourselves, and the definitions of these stories also limit our true experiences. When enough people believe in a new story, our culture will shift direction and bring about real changes. ## Stage 5: Cultivating a Garden of Harmonious Coexistence When I had a conversation with my friend who works in Google, talking about how different companies culture deal with the problems, I told him that I am a person always has a **best-case scenario** in my mind. He said: "That's why you are an entrepreneur." It doesn't mean I don't have risk management mindset. I just know every form of thinking is creative, but no thought is more powerful than the **original thought**. The process of creation begins with a thought, a concept, a visualization. All that we see around us was once a thought in someone's mind. I still remember how excited and joyful I was when I was doing the 3D modeling assignment in the 3D reconstruction class. I was amazed that I could create a whole new world by myself, even though I knew it wasn't real yet. We have all the tools needed to make choices. The world is what it is now due to the choices we've made. Our genuine thoughts are revealed through our daily decisions. My visualization of the future with AI is that there will be no shortage of goods and services anymore; it will be an age of abundance where everyone can have whatever they want. People can learn anything, anytime, anywhere, in their native language and in the way they like. We will have millions of AI bureaucrats everywhere, in the banks, in the governments, in the universities. Further, we will have universal basic income for everyone. I know some people might think this is utopian. But when we look back at the stage one in this article, human's desires are the beginning of all creation; it is the original thought. All needs should be treated equally, but not all needs are equal. It's a matter of proportion and balance. All we have to do is find a way to achieve harmony. So, what is your visualization of the future with AI? Share your thoughts on [Harmony Land](https://harmonylandai.vercel.app/). > Humans were tool builders, and we build tools that can dramatically amplify our human abilities. All inventions of humans, as history unfolds if we look back, are the most awesome tools that we ever invented. -- Steve Jobs]]> Tue, 01 Oct 2024 00:00:00 GMT https://www.hannieliu.com/mind/240928-ai-disruption Sat, 28 Sep 2024 00:00:00 GMT https://www.hannieliu.com/mind/240825-value-ai Sun, 25 Aug 2024 00:00:00 GMT https://www.hannieliu.com/mind/240808-three-films Besides Black Mirror, there are three movies released in 2023 have profoundly impacted my core values and beliefs about AI. You drop the bombs, and it falls on the just and the unjust. I don't wish the culmination of three centuries of physics to be a weapon of mass destruction. This line is the conversation between Oppenheimer and the chief science officer struck a chord with me. This echoes the idea that the impact of AI depends on its user, not the tool itself. Oppenheimer(2023) ## Barbie *To me, it’s a film about how a robot becomes human.* Barbie, a character who has long been stuck in repetitive daily tasks and scenes, starts to act differently without knowing why, even daring to mention the taboo word: death. Eventually, the feelings she's given allow her to experience rich emotions as she goes through various events, which I see as the key difference between AI and humans. Barbie(2023) ## The Creator *"Set them free"* The line that deeply moved me. Whether human or AI, when you can break free from labels, you can escape from the cage of confinement and gain freedom. The movie's vision of how to coexist harmoniously with AI or robots is also my vision. The Creator(2023)]]> Thu, 08 Aug 2024 00:00:00 GMT https://www.hannieliu.com/mind/231216-ai-ethics The world needs more entrepreneurs to think of the new ideas to bold enough to think of the new ideas and brave enough to pursue them. It's a call to action for all of us, especially women, to believe in ourselves and pursue our passions fearlessly.]]> Sat, 16 Dec 2023 00:00:00 GMT