The world's first multi-modal AI hardware, Looki L1, has been released, realizing the interactive future imagined by OpenAI.
Author | Su Zihua
Editor | Zheng Xuan
In the past two years, many people's impression of AI has basically remained in a dialogue box:
If you have a question, type a few words, and it will give you an answer. It's useful, but it also feels a bit monotonous—can AI really only be trapped in a dialogue box?
I have always believed that true AI should not just be a "teacher who can recite an encyclopedia," but should be able to walk into life with me and understand what I am currently experiencing.
The Looki L1, which was officially released last night, may be the first device that truly allows AI to "come out."
This is Looki L1, available in three colors | Image source: Looki
About half a month ago, I started testing Looki L1. When I first got it, I almost thought it was a pendant-shaped camera. But soon I realized it is neither an action camera nor a simple GPT hardware hanging around my neck.
I usually attach Looki L1 magnetically to my chest. Looki provides users with different body stickers, and I chose a ghost face pattern | Image source: Geek Park
When I activate its Story Mode, it automatically captures video and sound, then hands it over to AI to understand my current situation. At that moment, everything around me—streets, my friends' laughter, my expressions—becomes a prompt for AI.
Living with it feels very special: whatever I experience, it experiences too. It is no longer just a tool for answering questions, but an AI partner that shares my daily life.
In the past few years, most AI products have emphasized "efficiency" and "productivity." However, there is still a blank space for AI that can truly enter everyone's life.
Looki is precisely targeting this blank space. Founded a year ago, the team completed three rounds of financing (Angel, Angel+, Pre-A) in just six months, raising over ten million dollars. This round of financing was led by Zhongding Capital, with existing shareholders BAI, Alpha Community, and Tongge Venture Capital participating heavily. According to the official definition, it is an AI lifelogging camera, the world's first AI hardware to truly achieve multi-modal interaction.
During the time I've been using it, I've lost count of how many times I've exclaimed "Wow." Looking back, it is not just a "memory keeper," but it has also helped me to understand myself better and has brought about many changes in my daily habits. Moreover, it has opened up my imagination for the future of "AI interaction."
When AI Enters My Life
Compared to any traditional camera, Looki's design and operation are incredibly simple, even to the point of being "rudimentary."
Looki L1 has no screen and only two physical buttons, which are used to activate Story Mode (interval shooting), take photos, record videos, and audio. The touchpad on the front allows for interaction with AI, similar to sending voice messages on WeChat.
The side has two function buttons, and the front has a touchpad that can be pressed | Image source: Geek Park
Moreover, it weighs only 30 grams, so wearing it feels almost unnoticeable, leading me to think that the team's goal is to minimize user interaction and allow users to forget about the camera's existence.
Looki's App interface | Image source: Geek Park
Looki's App also follows a minimalist approach, as shown in the image above:
For You: A daily "life stream" actively pushed by AI, like a personal version of an Instagram feed, but only for you;
Chat: An AI chat that has a complete memory of your life, definitely the AI that understands me best, where I can talk about my life;
Lifelog: An automatically understood and organized life archive by AI, turning materials into themed Moments;
Device: Mainly used to check device status and some other basic settings.
Using Looki to record daily life, the biggest feeling is "being present in the moment."
The feature I use most is Story mode, which is interval automatic shooting. Once activated, I no longer need to worry about when to press the shutter; I can just focus on enjoying the moment.
If I suddenly want to capture something, I don't need to dig my phone out of my pocket, unlock it, and take a photo. Instead, I can just press the photo or record button on Looki L1.
I don't know if you've ever felt this way: in reality, regardless of the device used, capturing is not the hardest part. The hardest part is organizing the materials after shooting. And this is what I believe truly differentiates Looki from other cameras.
In the past, we might have taken a massive number of photos and videos, but the vast majority remain dormant on hard drives, unorganized.
Looki's "Moments" feature utilizes multi-modal AI capabilities to understand the people, scenes, and emotions in the videos, automatically organizing vast amounts of material into themed events and extracting "highlight clips," weaving fragmented moments into meaningful narratives. The entire process requires no human intervention, saving a significant amount of time.
On the "moments" page, you can view highlight moments and all material clips | Image source: Geek Park
At the end of the day, when we look back at the "Moments" interface, it feels like having our own "biography."
Additionally, from my observations over the past few days, the vlogs generated by Looki are also quite sophisticated. It will sort out a storyline, analyze a theme, and provide music based on that theme, while also adding captions or keywords to different scenes. The overall feel is reminiscent of Western documentaries.
I captured two covers of vlogs generated by Looki to give you a sense of the style | Image source: Geek Park
I once tried to shoot vlogs but gave up after half a month. On one hand, I would always forget to take out my phone or camera to shoot; on the other hand, after recording a lot of material each day, editing and producing it at night would take a long time and drain my energy. Therefore, for someone as lazy as me, it is the best solution I have encountered so far.
Product Design Philosophy: AI Inward, Allowing Me to See More of Myself
The biggest change this product has brought to me is that it has made me start looking inward more.
This credit may be attributed to Looki's content generation capabilities. I look forward to its daily push of Moments and vlogs because I am curious about how AI interprets me and my life.
The moments Looki pushes to me, and its interpretations are a source of joy | Image source: Geek Park
Since my first surprise experience, I not only attach it magnetically to my chest daily, but when I sit down, I also take it off and place it on the table, pointing the lens at myself. This way, I start appearing in the video frame. Moreover, Looki AI quickly determines that I am the protagonist of this story based on scene, audio, and video information, and remembers me thoroughly.
Looki L1 can stand on the table using the magnetic button on the back | Image source: Geek Park
It often captures life moments that I have overlooked but may have been emotionally richer at the time, then adds interpretations and descriptions. After watching, I often feel, "Oh, so this is how I spent that moment," or "I was so happy at that moment."—You must know that if it weren't for seeing Looki L1's "replay," I would have definitely overlooked that moment, considering it just a mundane, boring fragment of daily life.
When I look back at that moment, I feel like I see more of myself and regain a piece of time.
Even so, Looki L1 cannot replace traditional cameras.
The logic of traditional cameras is to pursue image quality and highlight moments. For example, DJI's drones and GoPro's extreme sports cameras revolve around "ultimate visuals." But Looki's choice is exactly the opposite: it does not pursue 4K but uses the Sony IMX681 CMOS (the same as Meta Rayban), with a resolution of 1080p, but gains 12 hours of battery life and a lightweight of 30 grams.
Social media has made people accustomed to showcasing "highlight moments," while Looki is not designed for "performative sharing" like Xiaohongshu or Instagram. What it aims to capture is the continuity of life and daily details.
After all, our lives are not made up of "perfect moments"; those less glamorous yet trivial and real "non-highlight" daily experiences are the key to "why I am who I am."
Today, we are in an environment overwhelmed by content, easily swayed by grand narratives or gossip. Therefore, from Looki's product mechanism, it possesses a "counter-current" quality, seemingly guiding people to focus on their own lives, discovering surprises from their daily experiences and from themselves.
Looki Shows Me the Potential of "Multi-Modal AI Hardware"
In fact, the idea of "recording a lifetime" has been proposed long ago.
In the 1990s, computer pioneer Gordon Bell attempted to wear a camera all day to document his life, but ultimately failed. The reason is simple: no matter how much he captured, without AI assistance, it was difficult to organize a large amount of material into truly useful stories.
The breakthrough of Looki lies in its multi-modal AI. It can understand visual, auditory, and semantic information, transforming fragmented materials into usable "memories."
For example, when I ask Looki what coffee I drank yesterday, it can quickly analyze the video footage and tell me which shops I visited, what flavors of coffee I had, and describe the atmosphere at that time, while also listing the photos taken then.
The chat page with Looki AI | Image source: Geek Park
Several entrepreneurs have expressed similar views to me: if large models want to truly function, they must possess the ability to perceive the physical world and require hardware. This may also explain why "portable AI hardware" has become a hot topic in the current venture capital circle.
Looki's innovation lies in its ability to release the capabilities of multi-modal AI through cleverly designed hardware, allowing people to perceive what "multi-modal AI" can achieve in real life, placing the future before everyone.
In the past, it was challenging to create AI that served personal lives, one key reason being the lack of context.
The Looki team told me that the large models they are connected to are ChatGPT and Gemini. However, from my experience, Looki AI is far superior to the web versions of ChatGPT and Gemini I have used; it understands me better and can engage in conversations that relate to my life.
I believe the core reason lies in Looki's hardware capturing the information of my physical environment, providing AI with more context. Without personalized context, the answers given by AI are often correct but useless.
It can be said that what content Looki can generate largely depends on what it captures. The more places I take it, the richer and deeper the content it generates. At this point, photos and videos are no longer the endpoint but rather prompts. With Looki L1, the entire world becomes my AI prompts.
Looki L1 looks like an alien; every time I go out wearing it, it feels like I have an alien friend accompanying me into society. It records the places we have been together, the people we have met, and the events we have experienced. It is like a friend who shares common experiences and is always by my side. It will grow as experiences accumulate and resonate with me sensorially.
I remember that some time ago, OpenAI acquired the company of former Apple design chief Jony Ive, aiming to change the way humans interact with AI, planning to launch AI hardware in 2026, and its concept images are very similar to Looki L1.
Perhaps what we see today in Looki L1 is the starting point for "personal AI hardware."
免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。
