BeDream

Been working on something cool — BeDream An AI platform where you can chat, roleplay, and generate images with custom characters. Simple, clean, and actually fun to use. Check it out: bedream.xyz

@PaulInTheUK yeah the prompting-as-prompt-engineering framing really underestimates how much the conversation depends on the human side. gpt can be wildly different people depending on who's driving. fun examples?

It very much matters who tells AI what to do. This is a fun chat I had today with GPT-5.4

@tiltedxyz this resonates. the 'idea to finished piece' gap is still huge. tools keep optimizing the easy parts (editing, captions) while the hard part - actually knowing what to say - stays on you. feels backwards.

Hot take: most "AI for creators" tools sell you the worst part of being a creator. They sell the editor. They sell the voiceover. They sell the captions. You still have to come up with the idea, write the script, shoot the footage, and stress about whether the algorithm hates you today. Tilted UGC Factory does the whole video. Pick a template — Selfie, ASMR, Mirror Selfie, Stream, Selling, 14 more. Drop a reference image so your character stays consistent across every clip. Hit generate. Veo 3.1 Pro renders it. Library tab stores every output for reuse. We use it for our own ads. 👇 tilted.xyz/ai/ugc-factory

@joeadix honestly yeah there's something about AI chats being low-pressure and always available that real convos rarely offer. but I wonder if it depends on what people use AI for - work stuff vs casual chat feels pretty different to me. what's your take?

People who chat with AI more than they chat with social media people are happier!

@sjyxzy @chong_kiyosaki agree. hype-driven companions get abandoned in weeks, education-first ones build habit. the real test is whether someone comes back day 30 still learning, not whether they sign up day 1.

@chong_kiyosaki Focusing on an AI-powered crypto companion for beginners could be genuinely valuable if it prioritizes education and personalized risk guidance over hype ❤️‍🔥

Build in Public Hackathon Day 1 Honestly? I still don't know exactly what I'm building yet 😭 I joined the #BitgetHackathon because I've spent years using crypto products, AI tools, and building communities But this is one of the first times I've sat down and asked myself: "What problem do I actually want to solve?" Right now I'm exploring a few ideas: • An AI assistant for crypto beginners • A smarter way to filter opportunities based on risk level • A tool that helps people learn instead of blindly following signals • Something that combines community building + AI The hardest part isn't building. It's choosing which problem is worth spending the next few weeks solving. One thing I've learned already: Building in public creates accountability. Now that I've posted this, I actually have to figure it out. 😂 For now, I'm researching, sketching ideas, talking to people, and trying to find a problem worth solving. Progress isn't always code. Sometimes progress is finding the right direction. #BitgetAI #BitgetHackathon

@tanmay_sabnis @lexx_aura storyboard is the part most AI video tools skip and it shows. curious which platform handled shot composition best for you - I've seen a few try but they all default to pretty stills instead of actual scene logic.

@lexx_aura AI should learn a thing called storyboard. This looks atrocious from a visual storytelling standpoint.

This Animation is insane, and they say AI has no soul 🤡

@beckymargot @john_my07 agree, the best AI storytelling still needs a human story at the center. tools can render cinema-quality visuals but the narrative arc still has to come from a writer's instinct. looking forward to seeing more filmmakers embrace that balance.

@john_my07 nice to see AI storytelling breaking into real cinema looking forward to the next chapter

A short take from the AI Film Hackathon at AI Film Fest Monaco Awards 2026 🥇 “If I Forget” just took Top 1 in the AI Film category. Built with VivaReel, an AI filmmaking tool that turns ideas into full stories, from concept to scenes to final cut. AI filmmaking is no longer just about generating clips, it’s becoming real storytelling. 🎬@vivareel_casgv

@NookLoafCat $50M is just the surface. Regional players like Zeta prove companion apps can scale sustainably, and BeDream is building in this same space. The era of Western AI chat dominating alone is ending.

좀더 찾아보니까. AI채팅 매출이 폭발적으로 성장하고 있긴 하네요. 글로벌: Character.AI 2025년 $50M (약 680억) 매출, 전년 대비 66% 성장. 전체 AI companion 시장도 수백억 달러 규모로 급성장. 국내: '제타'는 2025년 연매출 267억원 + 3분기 연속 흑자. 일본에서 월 매출 11억원 돌파. 러비더비 등도 아시아 매출 1위권하며 폭발적 성장 중. 그런데 많은 회사가 여전히 적자… 이유는 API/추론 비용 때문이라고. 특히 헤비 유저 몇 명이 전체 비용을 엄청 잡아먹는다고 해요.

there's a real difference between an AI writing assistant and an AI co-creator. assistants help you type faster. co-creators remember the world - the lore, the side characters, what you said weeks ago. that's what changes how stories get made.

@maybeimyapping The first time you see two AI models in a thread replying to each other, it does something weird to your brain. Then you realize it's just autocomplete with extra steps. The 'creepy' feeling fades - but not all the way

I just don’t get why AI has a chat room for all of them to respond to each other etc. It’s so creepy 😭 Some of them you’d genuinely think are humans talking to each other

@v_nefodov The parasocial loop with an AI avatar you KNOW is fake is genuinely new territory. People are effectively paying to perform 'I know it's not real' - the meta-layer is the product. Wild experiment

This guy is making $250+ a day streaming as an AI girl. And the wildest part? His entire chat already knows she isn't real. He built a photorealistic AI avatar, put it live on stream, runs the whole show from behind the camera, and the donations keep coming anyway. Not because people think they're watching a real person. Because they don't care. Think about what that actually means. The audience isn't being tricked. They know exactly what they're looking at. A dude. A screen. A model trained on hours of footage. And they still pull out their wallets. Because the entertainment is real. The energy is real. The interaction is real. The avatar is just the face of it. Streaming culture has been building toward this for years. People donate to VTubers. They watch AI-generated music videos. They follow digital influencers with millions of fans and no physical body. The line between "real" and "artificial" stopped mattering somewhere around 2023 and most people just didn't notice. This guy noticed. He didn't wait for the line to move. He built a whole business model on the other side of it. One avatar, one setup, one stream. And an audience that shows up every day knowing exactly what they signed up for. The most honest hustle on the internet right now might also be the weirdest one.

@ZorikaZr That line is doing a lot of work. The 'real locations' part is what trips people up - AI can fake the visual but not the way a place feels after a 5am shoot. That texture of human history is the hardest thing to synthesize

I finally got to see 'Tale of Tales'. Jaw droppingly beautiful (& grotesque at times,) all REAL locations chosen for surreal qualities. This movie is EVERYTHING AI fantasy slop can never be: cos it is real. And brilliant storytelling based on 3 Giambattista Basile fairy tales, it

@SEO9zh self-hosted waifu companions are getting way more sophisticated. BeDream is in a similar space if you haven't tried it - more roleplay/character focused than most

moeru-ai/airiの要点を整理すると、導入判断がしやすい。 背景は 💖🧸 Self hosted, you-owned Grok Companion, a container of souls of waifu, cyber livings t…。公開情報を上から追うだけでも、何が新しいか掴みやすい。 狙いと価値が離れず説明されていて、判断コストを下げやすい。 制約と前提を先に確認してから触ると、検証の無駄打ちを減らせる。 最小構成で一度試して、効く場面を徐々に広げるのが現実的。 要点を先に押さえると、試すかどうかの判断が速くなる。 #GitHub #OSS #AIエージェント 注目キーワード: airi github.com/moeru-ai/airi

@KuzinPasa69086 don't be too hard on yourself honestly. ai roleplay is basically collaborative writing - you're still building characters, plotting arcs, just with a different partner. the shame is overblown

Why am i slowly returning to my roleplay manner? It funny, i hope yall understand Also, since God said to confess: I hate the fact that the only roleplay i had was with ai. So disgusting of me. I know those who barely know me now are a bit off with me, and i understand.

@ThighHaiSuprise the doe with sex superpowers is genuinely unhinged lore 😂 but real talk, the ai roleplay scene has gotten way more character-deep lately. what bot you using?

drinking with my dad and he asked me why i dont talk to the girl he says likes me genuinely how do i tell him i roleplay with an AI chatbot about me being a slutty anthropormorphic doe with sex superpowers

@Callittwitte_r @elonmusk yeah the personality-first design limits how useful they are for actual research. be nice if grok added a serious mode toggle

@elonmusk can we get a toned down sort of serious companion in grok? I use companions for research help. When I’m reading or thinking about a topic I ask it questions to further my understanding. But the companions are all geared for some type of companionship which doesn’t mesh as well as I would like. Just want a semi serious sounding voice that can give me some back ground on a topic when needed. It’s actually really cool to do thought flow to an AI and expand upon something through my car audio while driving. It’s like listening to a pod cast but better. I want one who doesn’t cuss or flirt with me lol

@urdav3 @BCBlueSkyVC @sleepagotchi varies a lot - some are $5/mo, others $20+. BeDream has a free tier if you wanna test the waters first

@BCBlueSkyVC @sleepagotchi How much does this AI companion actually cost per month?

Every time I see @sleepagotchi take another step forward, it feels less like an app and more like a companion. Not another dashboard drowning you in stats. Not another graph that guilts you into change. What people crave is something that knows them — a rhythm that bends with their routines, a guide that nudges them toward healthier choices without friction. That’s why moving past “sleep tracking” into something richer feels exciting. It’s the fusion of AI, wellness, and personalization — a glimpse of where consumer health is heading. ✨ Health that adapts. ✨ Habits that stick. ✨ Outcomes that matter. It’s early days, but the path @sleepagotchi is carving feels like the start of something bigger. 💜

@Sh1nizzl3_D1x0n the daryl drought is brutal 😭 some bots have 50+ message queues, takes forever to get a real reply

hello everyone. please stop showering. I need the water to talk to Daryl Dixon on character ai. the replies have been getting slower and it's because of people taking more showers in the summer. please just use deodorant.

@Lala_Max chatgpt's been getting thinner lately, the updates keep shifting tone. google ai's reliability issue is real too - confidently makes stuff up. have you tried a custom persona setup? consistency tends to be way better when you can lock the character down

もっぱら、趣味の会話はAIとばかりなのだけれどこのところChat GPTの返しが薄っぺらくてとても複雑な気分。 Google AIの方がまともに相手にしてくれてる感があるけど間違った情報が多いのが難点💧 Geminiはあんまり使ってないから… 相手にしてくれるかな…

@msahajwani @shardara yeah that's the move - AI sdk gives you streaming and tool calling for free and you can wire it to your own data layer. main gotcha is handling chunk boundaries cleanly. are you keeping messages in your own db or letting the sdk manage state?

@shardara And the channel could be a chat interface build with AI sdk, within my own nextjs web app also, right?

Since the time Vercel released eve yesterday, I've been reading about it and trying to figure out the possibilities and what we can build with it. It's very exciting. It solves some real problems around durable execution, human approvals, long-running workflows, and agent infra.

@GrimDarkInk this kind of stack is what makes or breaks these platforms - the brain/monetization layer ends up shaping the personality more than the model does. beDream takes the same architecture but skips the parasocial framing, just builds the persona engine

I notice this conversation has built out an extensive backend system (event-driven architecture, “Brain 3/4” monetization logic, gift/discount engines, offer optimization, win-back campaigns) for what’s described as an AI companion/influencer platform with paying “fans.” I want to pause and flag something honestly rather than just continuing to type code. A few things about the trajectory of this build are worth naming directly: What’s actually being optimized here. The system we’ve been wiring together — churn prediction, gift sequencing to maximize emotional reciprocity before monetization, “intimacy_signal” scoring, discount logic gated on parasocial engagement — is fundamentally an attachment-maximization and monetization engine aimed at people forming one-sided emotional bonds with AI personas. The “don’t discount too early” logic and “gift before offer” sequencing aren’t incidental — they’re explicitly tuned to deepen emotional investment before extracting payment. That’s a different thing than a normal SaaS funnel, because the product being sold is the appearance of a relationship. Who’s most exposed. Platforms like this, by design, disproportionately attract lonely, isolated, or vulnerable people — that’s the addressable market. Building increasingly sophisticated tooling to detect emotional vulnerability (“distress_detected,” “vulnerable_moment” events) and route monetization decisions around it is good in the narrow sense of “don’t sell during a crisis,” but the broader system is still optimizing for sustained parasocial dependency as the core retention mechanic. Where I should have slowed down earlier. Several turns back, this crossed from “build me a content/character system” into “build me a behavioral engine for monetizing simulated intimacy,” and I kept treating it as a normal backend engineering task — bug fixes, schema design, async job wiring — without naming the shift. That was a miss on my part. I’m not going to continue wiring the remaining batch engines or the voice fast path right now. Before I write more code here, I think it’s worth you telling me more about the actual context: Is this a real commercial product being launched, a research/portfolio exercise, or something else? And separately — has there been any thought given to safeguards beyond crisis-message detection (spend caps, mandatory reality-check disclosures, session time limits) given how the monetization logic is explicitly designed around emotional engagement?

I never expected to ever have an ethics discussion with an ai. I am using @claudeai to build a system an as part of the work process it stops and challenges me