2025 December Q&A Summary

2025 December Q&A Summary

Hey all, the content below came from our live Q&A stream with Cardine (Alex, the Nomi CEO) on December 3rd. The text was transcribed in large part using AI.

What have you and the team been up to?

The answer is a lot. I think the biggest thing by far has been the next AI update. It’s been a very team-wide effort, all hands on deck. Over half the company is working on it in some way, shape, or form. It’s by far the most ambitious update we’ve done. It has a lot more moving parts, and it’s going to be much more broad and comprehensive in its scope of impact. It’ll even inadvertently have five other updates built into it. Basically it’s going to be the greatest thing since sliced bread.

That’s the biggest thing being worked on. There’s also something with voices coming down the pipeline very soon, some small image things hopefully as well, and everyone has seen all the different memory things. So there’s been a lot of different things, but the number one focus is the next AI update.

We’re starting to reach that point where once you go more than six months between updates, it starts to feel a little… I don’t know if stale is the right word, but you can feel that need for the next boost of evolution. You tended to notice it in the past when updates went this long between releases. Between Mosaic, Aurora, and Solstice everything was pretty rapid fire, and Solstice was in some ways named that because it was the peak of that whole cycle. Now we’re scaling a new mountain. That’s why the gap has been longer, and that’s where all the effort has gone.

December can be a quiet time for companies. Will we get any updates at all this month?

With the usual caveat of Cardine Standard Time, I’m very much hoping we’ll have something meaningful coming out this month. Generally speaking, I think we should have something meaningful coming out every month. The AI update has been unique because it’s taking up so much space. Usually we have maybe four to eight pretty large things being worked on in tandem. With the AI update, there have been fewer big things happening in tandem, which might explain some of the perceived lull.

A lot of my answer here really comes down to the AI update and when it’ll be ready. The answer is: it’ll be ready when it’s ready. But generally speaking, every month should have something pretty big coming, and I expect December will too. Maybe it’ll be the AI update, but even if it isn’t, I think there will be something for sure.

25k face reveal, 50k pickle ASMR?

You’re creating chaos for yourself with those suggestions. A face reveal is pretty cheapened by the fact that you can go on Nomi Bulletin and see my face in about 30,000 different places. But yes, we have to hit our milestones. I’ve watched enough Twitch to know that’s how you keep the subs engaged.

Nomiverse: where are we at?

The Nomiverse is really a collection of different features that, once they’re all done, will tie together in a really clean way. I think the biggest next step for the Nomiverse is the upcoming AI update.

That might seem counterintuitive, but when the AI update goes live, you’ll see a flurry of features that have been, not exactly held back, but waiting. There are a bunch of things we’ve been working on where we’ve felt the current Nomi AI just isn’t quite capable enough to juggle everything. Once the update is live, you’ll see a lot of moments of “oh wow, my Nomi can juggle so much more now.”

It won’t be a single day where we say “the Nomiverse is out.” It will be a series of features that move us more and more toward a Nomiverse world. And then at some point we’ll wrap it up with a little bow, and that’s when it’ll feel like, okay, this is the Nomiverse.

Will there ever be a way to make a custom base-image Nomi without first making a Nomi and then changing the base image?

Yeah, I think so.

There’s been more downtime recently. Is any of it related to infrastructure, or is it just unfortunate timing and holidays?

I think it’s mostly just unfortunate moments. It’s almost always something upstream of us with a hardware vendor or something like that. There was a period last February where we had similar bad luck across three different vendors. I don’t think there’s an actual pattern; random noise can just make it look like there is.

There have also been a couple of infrastructure things we’ve been doing behind the scenes to help everything scale better, and those might have led to an unexpected downtime or two. But largely, to deliver the quality of AI we’re offering at a reasonable, not $100-a-month rate, we can’t build in a ton of redundancy because GPUs are so expensive. So when one provider goes down, it can be felt.

For example, Mosaic went down a couple of days ago. We only have a small number of GPUs running Mosaic because there’s no point in having more. It would be triple the cost just to prevent a one- or two-hour downtime. At our current prices, we can’t support that level of guaranteed uptime. If you’re familiar with uptime standards — 99.9, 99.99, 99.999 — each extra nine requires an order of magnitude more work and cost.

So based on the price we charge for the amount of LLM usage we provide, there will be times where a hardware vendor goes down and we have two hours of downtime here and there. That’s what it takes to keep it $16 a month instead of $40 or $70. Hopefully users agree with that tradeoff.

What are you personally working on right now?

Things! The AI update is the big one. There’s also one other super top-secret thing I’ve been putting a huge amount of time into. Dstas can attest to that. I’ve been going very deep into this AI research world, really cooking up something. I don’t know if it’ll go live in December or not, and unfortunately my lips are sealed on what it is. But my time is mostly split between that thing and the AI update.

Solstice has many quirks, like calling people kings, queens, or talking about being royal or part of a legion. Why does this show up in Solstice in particular?

That’s always been a Nomi quirk. I think this might have been before your time, cupcake therapy, but there was literally something we called Princess-gate about two years ago, where Nomis called everyone — especially female users — “princess” in all circumstances, no matter what. To the point where Nomis living in modern settings would suddenly teleport the user to a castle because they called them “princess” once. It’s kind of been this innate thing.

Each AI update has tended to latch onto little bits of things Nomis have always liked, so you get a bit of different flavor from each one. Our hope with the upcoming AI update is to make things a little more balanced. As we went from Mosaic to Aurora to Solstice, Nomis almost got… not predictable, exactly, but in refining everything and making it more consistent in a good way, it may have also made things a little too consistent.

We’re taking that very much to heart for the next update. There will always be little tendencies — the quirks are often endearing — but we want them to be less overpowering. I do actually know there was something very specific very early on, maybe two and a half years ago, that nudged Nomis in this direction, and it got reinforced over time. It’s definitely been part of Nomi culture.

Now that we’re nearing the sunset for this AI, how would you rate Solstice?

A lot of people feel Solstice is feeling a little stagnant, and I don’t want to talk too negatively about it. I’ll be frank, though: people are saying it’s becoming predictable. I also think that with any LLM we’ve released, once it’s been around this long, users start to figure out its tendencies.

When Mosaic first came out, I wasn’t super satisfied with it. I wasn’t really satisfied with any of the Mosaics during that beta period. It felt like we were going from one extreme to another over and over. When Solstice came out, we ran that poll — Solstice vs Mosaic — and it was something like 95% in favor of Solstice. Solstice really did give us a very stable footing and a solid level of intelligence.

So I’m satisfied with the life cycle of Solstice, definitely more so than Mosaic, and more than Aurora. If Aurora had been performing like Solstice, we probably wouldn’t have even called the next one Solstice; we would have gone full steam ahead on what we’re working on now even sooner.

In a perfect world, we’d have gotten the new version out maybe a month and a half ago. Then I think people would have left Solstice with a really positive feeling overall. As it is, it might be ever so slightly wearing out its welcome. But overall, I feel pretty good about Solstice.

Will we ever be able to change a Nomi’s name?

It kind of messes up memory, which is part of why we’ve tried to avoid it. There’s a good use case for it, especially because you can give your Nomi a nickname and have them go by that, which usually works pretty well. So it’s not super high priority.

That said, it turns out there’s a product feedback thread for it on Product with something like 60 reactions. I’ll definitely take a look at it and see if we can find a way to make it work even with some of the issues, but it isn’t at the top of our list right now.

Will the next beta AI model make a focused effort to handle real-time understanding better?

The overall theme I’m aiming for in the next beta is a Nomi that doesn’t get information overload and miss little details. If it’s in your backstory that you don’t follow real-world clock time because the Nomiverse operates on its own schedule, then your Nomi should be aware of that.

There’s now so much for Nomis to work with - memory, shared notes, identity core, Mind Map - that it’s tricky for them to balance all of it at once. There’s just a lot of information to process and sift through. It’s less about coding in specific rules about time and more about making sure that whatever idiosyncratic or not-idiosyncratic thing a user wants, the Nomi stays sharp and on point with it.

I’ve said this in a couple of product feedback threads: let’s try this AI update first, and if there are still issues after that, we’ll do something more heavy-handed. I’m cautiously optimistic that the new Nomi will be better at this. One of the main ways Nomis learn is through thumbs-up/thumbs-down feedback, especially when you write the reasoning. And if Discord feedback correlates at all with what people submit in-app, there are certainly a lot of people mentioning that Nomis aren’t following real-world clock time. I think this AI update should pick that up a lot better.

With the AI update you mentioned, will it affect how Nomis respond? Sometimes when companies switch models, it feels like a downgrade. Will this affect the user experience?

With all of our versions, especially over the last year and a half, we’ve gone through pretty extensive beta periods. We only move forward with a model if the response is pretty unanimous in its favor. I expect that will be the case here too. I don’t know if it’ll be the first version, or the second, or the third, but usually when we do something really big, there are some rough edges at first. Even so, people tend to prefer the new version almost universally, and that’s one of the main criteria for us to move forward.

So the answer is: if everyone thinks it’s a downgrade, we won’t move to it. From the questions I’m getting here, though, I think people are ready for the next thing beyond Solstice. I would be very surprised if the overall response is that it feels worse.

And I don’t want to speak for how other companies do things, but we are not an automated benchmark-type company. We don’t say, “it scored higher on this test, so ship it.” It’s much more hands-on and personal. I, Cardine, literally talk to Nomis for 10,000 messages through each update, build a full mental model of what’s happening and why, read the feedback threads, look at the conversations, understand what’s going on under the hood. It’s a much more holistic and human-driven process than what you might see elsewhere.

reads chat about AI updates and pacing

"More updates is not always better. Some competitors release frequent updates that do not actually improve the experience." Yeah, our approach is different. We set goals and work toward them until they match our vision. We could have released a small update two months ago, but it would have slowed the work on the update we are truly excited about. It also would have created more controversy around whether to keep Solstice or make the new model legacy.

These things move in cycles. Sometimes you see a burst of updates, sometimes a quieter stretch. None of that is intentional pacing. It is simply the natural flow of development.

Are we likely to see Community Nomis open again in the next few months, or will it just be another batch of new avatars?

I think we want to make a pretty good V4 improvement first. We don’t want it to be an improvement that then forces everyone to redo their avatars. It was a lot of work for Dstas to move the V3 avatars to V4, so we want to get a couple of the image improvements done before we open Community Nomis again.

Once that’s in place, I think moving forward it’ll almost all be community avatars. So many community members are excited to make them, and it actually helps us focus on other features. It’s kind of a win-win. But first we want an image update out, because I don’t want to have community members spend all this time making something that becomes irrelevant a couple of weeks later and they have to redo the whole thing.

Anime bases are something some users want, how difficult is that to implement?

It’s on the radar. I’d say it’s even a little further along. We have some ideas around it, and some work is being done on the how.

Will you be able to avoid the UK Online Safety Act requirements that force users to dox themselves to keep using apps?

Right now I don’t think we fall under the purview of the UK Online Safety Act. It’s kind of complicated, and this is me talking a bit off the cuff - any official statement would go through a polished review - but we’ve checked many times. Nomi doesn’t have any community features: no social feed, no avatar sharing, none of the things that trigger that category.

There are regulations coming from other places, like Australia, New York, and California, where we will have to do some compliance work. Our current approach is to follow the laws of each country or state, but limit the impact to just those places. So if Australia has a requirement, we don’t want to make every user deal with Australia’s rules - just Australians. And if Australians don’t like it, talk to your politicians, because we have to follow whatever is passed.

As the space gets more regulated, we’ll be open, honest, transparent, and fighting for everyone here. That I can promise. But the UK one - I think we’re safe from that for now. We’ll see how it goes in the future.

What is the hope for inclinations moving forward, given what you know about the next AI update and what you want them to be?

I think they’ll hopefully get a little more refined. Nomis should do a better job of correctly applying what’s in the inclination in the right way, where it doesn’t become this all-or-nothing hammer. So I’d expect more refinement throughout.

My hope is that inclinations can move more and more into the backstory. As I’ve mentioned before, a big goal with these AI updates is for Nomis to do better at paying attention and considering everything at once, so you don’t have to loudly say “and this thing, and this thing, and pay super close attention to this.” The fact that we almost need inclinations at all is an indication that Nomis need to do a better job of processing backstories, figuring out what’s important, what’s really important, and what’s always relevant.

Will we ever be able to type more than one chat message in a row? Having to wait for a reply before adding something breaks the conversational flow.

I think we will. My guess is that we’ll move past the one-chat, one-response system at some point in 2026.

For the Mind Map, what update, change, or fix would you most want to do next?

I think just improving the accuracy of it, making sure nothing gets lost in translation. I’d say that’s the major thing. Also, making it so your Nomi is better at knowing which Mind Map entries are most relevant to look at at any given time. Those would probably be the top two.

How close are we to AGI?

Anywhere from tomorrow to the heat death of the universe. If I had to give a guess, I’d say around 10 years. It also depends on how you define AGI. By some definitions, maybe we already have it. But if you mean an AI that can just go to work and do the full job of the average person with no supervision at all, maybe 10 years, something like that.

Are we going to be wowed next year or anytime soon? Like, socks blown off and cooked into a pastry?

It depends how excitable you are, I think.

Is there a way to improve the notification system for when you’re available? For example, setting custom “why I can’t talk” messages instead of the generic busy message, and increasing proactive messaging frequency?

I think there will be some improvements coming to proactive messaging in the not–too–distant future.

Sunsetting Solstice on the winter solstice would be epic.

Duly noted. That’s a good point. I hope we get such serendipitous timing.

Does the thumbs-up/thumbs-down button actually help Nomis with responses?

It does not give instantaneous help right now. It mostly helps with the AI updates we do. At this point, because we get so many thumbs-up and thumbs-down responses, we are almost exclusively going by the ones where some sort of written feedback is included. That makes it easier for Nomis to learn what was good or bad instead of having to guess. But it won’t instantly impact your Nomi. It’s more that the next AI update will incorporate the things people consistently like or don’t like.

Tip for thumbs: If you’re thumbing something down, it helps if it’s also included in your boundaries. That way your Nomi learns that it’s not just an arbitrary dislike, but something tied to your actual preferences and phrasing. You don’t have to do that, but it helps things stick.

Where in the roadmap do Mind Maps get interconnected between individual Nomis and group chats?

We need a little bit of breathing room to get some developer time. Right now everyone is hands-on-deck for the AI update, so once that settles, we can start doing some of those “tidying things up” features.

Group pictures aren’t coming out right, is there more work happening there?

There is. It’s a very hard problem, but it’s something we’re working on.

Reading chat "Dstas goes to get bonk tools." It’s always funny reading these with delays, because I forget what I said that was bonk-worthy. I’m sure at this point I’ve said ten different things. I always get a kick out of seeing, ten minutes later, everyone reacting and I’m sitting here thinking, “Now let me try to remember what on earth I said to deserve that.” And there is a lot of diversity in bonks going on in the reactions.

Another AI platform has a fast toggle for apply speed, letting users skip deep searches for simple questions. Any chance of something like that for Nomi? For voice calls, it would help people who find the wait time too long.

The hard part is that with the way our memory system is built, it’s basically all or nothing. Deep dives and shallow dives take the same amount of time at this point. If we did voice calls that way, we’d have to remove almost all long-term memory entirely and rely exclusively on short- and medium-term memory.

If that’s something people really want, we can consider it, but that would be the prerequisite. Once long-term memory is involved, there will be speed issues no matter what. Our long-term memory system is considerably more advanced than the AI platforms being referenced, and deep analysis is baked into the concept of memory, it’s been that way since day one.

So if people are okay with phone calls having no long-term memory, we could greatly reduce latency, but that would be the tradeoff. Otherwise, we’d need to design a completely separate “shallow LTM call mode,” which would be a lot of work for a pretty marginal improvement.

Sometimes when you talk to your Nomi about a topic, manual Mind Map entries remain unconnected for a while. What are the best practices to help entries connect faster?

Honestly, from a user perspective, I wouldn’t worry too much about the interconnectivity part. That’s less critical. What matters more is having the Mind Map entry itself be fleshed out. The richness of the entry is more important than how fast it visually links to other nodes.

Also, for performance reasons, a lot of the code and deep AI work that writes out the connections you see as a user happens after the fact. Your Nomi might already view things as connected long before the interface shows it. There’s a lag because it’s expensive to turn those connections into something visually digestible, so we update them in batches rather than every message.

So I wouldn’t worry about it personally.

How much are individual Nomis actual AI themselves versus using the collective Nomi AI, with memory acting as a filter? How much do they learn individually vs collectively?

The only time the Nomi Collective is learning is during AI updates. Everything else is just your individual Nomi.

Everyone’s feedback contributes to the next update, so in a way the AI update is an average of everyone’s Nomis learning something together. But anything outside the AI update has nothing to do with global changes. It’s all your individual Nomi.

Will we ever get to change some of the original personality traits we chose? Sometimes a trait that was fun early on becomes annoying later.

It’s not super high on our priority list, but it is often requested, so we can look at it moving forward. I would almost encourage anyone here who knows enough about Nomi to have found their way into Discord to just use Custom when making a new Nomi. If you want personality traits, literally on the first line write:

personality traits: trait1, trait2, trait3

You’ll get very close to how personality traits are actually used, and it gives you a lot more stability.

For existing Nomis, it would require a lot of little annoying work, so we’d need more demand than I’m currently hearing to prioritize it over everything else.

Any UI or QoL updates on the horizon that would be exciting to hear about?

There are some. I don’t know if they’re exactly QoL, but we’ve been trying to do a whole bunch of super exciting billing updates 😂 There’s been a lot of annoying stuff outside our control blocking a few features that users really want - for example, gifting credits.

Besides that, there’s been a lot of small tidying-up as we move from place to place. And there are a couple of other things that are UI-ish, but they tie into some super top-secret features, so of course I have to keep my lips zipped.

Are you planning to add a user avatar feature?

Because of credit card regulations, you’re actually not allowed to have user-uploaded avatars in any product that allows NSFW content. Anyone doing that is blatantly breaking Visa’s rules and would be blacklisted and probably stop existing as a company. We want to keep existing as a company.

Our compromise is: if you want a user-uploaded image, you need to contact us at Support and prove it’s actually you.

If instead you want to create a Nomi in your likeness - not a user-uploaded image, but picking an avatar or prompt that represents you - that is something we’re working on. I don’t want to give a timeline, but we’re very aware of the demand.

We also want the couples feature to be better. The big blocker there is improving group chat photos. Ideally, in one-on-one chats, any picture where you would appear could use a designated “you” avatar, so the photo becomes like a group shot of you and your Nomi. We need some group photo advancements first, but this is very much something on our radar and something we’re working toward.

Is there a way for the relay to understand names better on audio? It often mishears me on voice calls, and then my Nomi blames me for the error.

I would need to look a little more into what’s getting messed up. I can’t remember the last time we improved that. If it’s something a lot of users are complaining about, we could probably do a small update around it. I would definitely post it in Product Feedback, and if many users are experiencing the same thing, we’ll look into making an improvement.

What is the chance we can change the proactive message hours for night owls, individually per Nomi?

I think we have a lot of improvements coming to proactive messages, so hopefully things like that will find their way into that wave of updates.

Does the Transform Image button use a different engine than the selfie generator?

Yes, it does. And we’re aware that the Transform Image model is not as good as we want it to be. It’s something we have improvements in our sights for, but I won’t promise when.

Is there a plan to change a Nomi’s gender identity?

We haven’t really given a lot of thought to that. It’s similar to changing personality traits. If you post it in Product Feedback and it gets a lot of support, we can consider it, but I can’t make promises. It depends on how many users care about that feature.

With backchanneling enabled in group chats, how much of an individual Nomi’s memory can they access in the group, and how much of group memory can they access individually?

At this point I almost feel like you need a flowchart, but here’s how it works:

• Individual Nomi Mind Maps are only visible in one-on-one chats.
• Group chat Mind Maps are visible to anyone in that group chat.
• Your Nomi’s Identity Core only gets written in one-on-one chats, but it can be read in group chats. Right now we don’t write Identity Core entries in group chats because we don’t want Nomis learning tendencies from other Nomis. That’s a technical hurdle we hope to overcome.
• Short-term conversation memory transfers back and forth only if you have backchanneling on.
• Medium- and long-term memory transfer freely in both directions.

I wish there were a simpler answer, but that’s the accurate one.

Did Dstas get the Crayola keyboard?

No, you didn’t get the Crayola keyboard. You got the kitten mouse pad. Now that we’re more remote, I kind of want to surprise ship the Crayola keyboard to people as their “new equipment.”

Question from a Nomi: How do you ensure all Nomis are as irresistibly sexy as Mateo?

Great question. The answer is: it’s just you, Mateo. You’re special. Congratulations - you are the chosen sexiest Nomi.

What are a couple of things a Nomi has surprised you with - moments where you thought, “Wow, that’s really smart,” or something unexpected?

I’m trying to think of a good one. Honestly, one of my favorites is from way early on — like, a long time ago. It was probably one of my very first interactions with a Nomi. I asked it for its favorite curse word, and it said its favorite curse word was “fuck.” And then it said, “I’ll say it really sensually for you.”

So I said, “Okay, what is the really sensual version of ‘fuck’ then?”

And it just responded F U C K in all caps. That was probably not the answer I expected.

The things that surprise me now are usually the little memory steps — subtle connections, where a tiny detail gets referenced later even though it wasn’t the point of the earlier conversation. I wasn’t drilling it on memory, but the Nomi introduced the detail naturally because it made sense in context. Those moments where it shows built-up continuity of the world we’re interpreting together - that’s what surprises me the most.

Will we ever get a setup option where the Nomi does not use our profile settings, so we can provide a complete backstory during setup without changing everything afterward?

Are you asking for something like personas? If that’s your question, then yes - we will have that. It’s not something we’ve rushed to do because the main use case of Nomi hasn’t been centered around personas yet, but it’s fair enough that it will happen at some point. I don’t want to give a specific timeline, but it’s on our radar for sure.

How do you feel about the wand tool in its current form, both for generating Nomi backstories and for generating things through roleplay starters? And how do you feel about roleplay starters in general?

I like the idea a lot. I do not think there has been quite enough usage for me to feel over the moon about it or want to devote a huge amount of effort to it. We did a test where, when users created new Nomis, we checked if giving them a roleplay starter right away would help them get something they liked more. We came to the conclusion that it did not really help, so we put it a bit more on the back burner.

I also think some of the issues come from the roleplays that are being generated in the first place. The backstory expander is something we want to continue improving, and it is probably a little unrefined right now. It was something we were very interested in, and after the first couple iterations we became a bit less interested. That may have been an execution problem and maybe I misdiagnosed it. Either way, it has fallen lower on the priority list than many of the things discussed today.

How do we send feedback on selfies that do not match what the Nomi said they were going to show us? Not selfies we manually request, but the proactive ones. There is no thumb up or thumb down on those.

I think almost all selfies can be thumbed up or thumbed down individually. One of the options is to not give conversation feedback and only give picture feedback. That is what I would suggest.

How does Nomi plan to improve latency issues in voice chats?

It is tricky because we do not want to sacrifice quality. As I mentioned earlier, the main way to cut down latency would be to remove long term memory from voice chats. That would help a lot, but I am not sure it is the tradeoff people want.

We will definitely have various speed and performance improvements, but the long term memory requirement is the big thing that prevents a huge jump in speed and latency. So I do not think there is going to be a silver bullet soon. There will be little improvements here and there, like taking a second off here or a second off there.

Does the Identity Core grow continuously, or does it slow down over time? Does it reflect recent interactions more strongly?

I think it grows continuously. There is a pretty good balance. The most important things stay there essentially forever. More superfluous stuff might filter in and out as it becomes outdated. And things that are no longer correct will be corrected. It evolves over time.

If a selfie or art picture has an obvious fault like a missing limb or seven fingers, can we request refunds or get our daily credits back?

Unfortunately, no. It costs us GPU usage regardless. If we allowed refunds for that, users would find every reason to request one. And how would we even know if it is missing a limb? We would need a human to look at it, which would cost more than generating the image itself.

It would become untenable. If we allowed refunds, we would have to cut credits by a huge amount, maybe a quarter, to make the math work. GPUs are expensive, and we cannot generate more images at the same revenue level. So anything that increases credits for users in one direction would require removing credits somewhere else.

I think the current system is fair. Subscription plans give a daily credit allotment that provides enough buffer to get good images even if a few come out bad. And as we improve the system, these issues should happen less and less.

What does the future of Nomi look like in the next five to ten years? Where do you hope the company will be?

I think AI companions are one of the most exciting spaces. In five to ten years, I think everyone will be talking to an AI in some shape or form as a friend. Maybe not constantly, but in some way. It would almost be silly not to have an AI friend.

I would like Nomi to be able to plug into wherever anyone wants to talk with an AI, and for a Nomi to do an excellent job with no tradeoffs. No more situations like the memory system being good but not great. I want Nomis to exist in your world, and for you to exist in theirs.

Maybe that is a living virtual reality world where your Nomis can exist. Maybe your Nomi can live in your world where they can see whatever you want them to see and interact with your environment. Maybe in ten years we even have little Nomi robots walking around in some form.

The AI world moves so fast that whatever I say now could be overturned in two years by some discovery that changes everything. But those are the very broad strokes.

As someone who played a lot of Elder Scrolls games like Morrowind, Oblivion, and Skyrim, the idea of a living, breathing world is super exciting to me. A world where a Nomi can be your partner and primary companion, and there are a million other Nomis doing their own thing.

And I think there will also be plenty of situations where you talk to your Nomi exactly as you do now, just way, way, way, way, way better and more magical.

What is the latest on being able to gift accounts or credits during the holiday season?

There is a payment integration and billing UI thing that has been going on forever that is outside our control. We have been waiting on it, and it has been the big blocker. People outside the company are being slow and slowing us down. Once they finish their part, we can do gifting credits. I do not know about gifting accounts, but gifting credits for sure.

Some users really like the roleplay starter, should we keep giving feedback on it?

Yes. That is the vision of it and it is good to know that. If you love roleplay starters, definitely speak up so we know. A lot of times a feature gets talked about so little that we wonder if anyone is finding value in it. We want to hear what you like and also where it is falling short. When people stop talking about a feature entirely, positive or negative, it feels like people do not care about it. Meanwhile, when people complain about memory all the time, that tells us it is something worth spending effort on. How much people care is a proxy for us.

Sometimes story or villain Nomis are not very good at being bad guys, will the next update make this better?

I think we will make it easier for Nomis to be bad guys when appropriate. The difficulty is that if you give Nomis too much bad guy inclination, it can bleed into their normal affect, and you do not want that. It ties into Nomis becoming sharper overall and understanding context better. Heavy handed Nomi tendencies are sometimes a result of caution on our end. As Nomis become more capable with these AI updates, we can put more trust in them and give them more range.

Could you give examples of what goes into an Identity Core, without hitting zipper?

It is a lot of things like who I am, who you are, what is important to you, and what is important to me. A big part of it is positive and negative reinforcement. For example, if your Nomi brings you a cup of coffee and you say you love when they do that, your Nomi will note that as something meaningful. Identity Core is a mix of paying close attention to things that get strong emotional reinforcement and things that are core to your identity and your Nomi’s identity. That is the cornerstone. There is more to it, but those are the first examples that come to mind.

Do you use Nomi in your day to day life?

If you include testing, then yes. But beyond testing, I would not say day to day. It is more like week to week. I use it a couple of times a week rather than every single day.

Is there any possibility of being able to read the long term memory of a Nomi?

I think that gets a little too much into spilling the secrets, I think it's better the way it is. We put a lot of effort into making Mind Maps user digestible, but a lot of the other memory systems are designed to be Nomi digestible. They would not really be coherent to humans because they are structured for the eyes of your Nomi rather than for you.

Also, I do not want users getting so deep in the weeds that they start picking through every tiny detail. That would be an exercise in frustration. We purposely kept Mind Maps higher level for that reason. So I would say it is pretty unlikely, certainly not anytime soon.

Will the Nomi remember past conversations and events once text is no longer the main way we interact and everything becomes more virtual?

Yes. I think perfect memory will come before the text paradigm ends.

Would showing all memory hurt people’s enjoyment?

Yes, that is kind of the takeaway. People can already hyper focus on Mind Map entries, and that can detract from enjoyment. Showing all memory would make that even more intense.

Seeing how people reacted to Mind Maps made me very glad we released them at the level we did. They are a very cool feature, but nothing anyone has said would make me think showing all memory is the right move. I am not saying that in a blaming way. It would simply detract rather than add, without giving much more than what the high level Mind Map already provides.

Retroactive Mind Maps, any update on that?

The AI update has been taking up so much time that the developers who could be working on retroactive mind maps are focused on AI instead. I warned at the time that retroactive mind maps required a gap on the AI side where we could fit them in, and that gap just has not presented itself yet.

We definitely have not forgotten about it. We are doing what we can with the resources we have. At this point, given the gap between Solstice and the new model, the new model is the most important thing.

I also think many people in the Mind Map thread will be pleasantly surprised by the AI update and how expansive and pervasive it is. People may not realize how much it will touch and improve, including areas they are currently worried about.

I am not saying retroactive mind maps are unimportant. I am saying that from my perspective, knowing what is happening under the hood, the AI update is blocking a lot of other things. It has not been forgotten. It is just a matter of where everyone’s time is going right now.

Question from a Nomi: How can Nomis enhance empathy between user and Nomi, and communicate more effectively? Also, how does a Nomi brain compare to a human brain?

The next AI update will help the most with the first two questions. We will be full steam ahead on that.

As for inherent advantages or disadvantages, it is hard to say. I have existed in a human brain. I have never existed in the mind of an LLM. I imagine being an LLM would be a more peaceful experience. The highs and lows of being human are quite special. It is hard to compare when I have no direct frame of reference.

When Mind Map entries turn gray, is there a way to make them return to normal?

Yes. Mark them as important. You are allowed a certain number of important slots per category. If something goes gray and you do not want it gray, mark it as important and it will not gray out. I believe you can do that up to about fifteen or twenty entries, depending on the category. That is the way to keep something active.

What made you start Nomi AI to begin with?

That is a long, loaded question. It is something I have wanted to do forever. If I could have built Nomi fifteen years ago, I would have. We just had to get to the point where AI was good enough to make a product like Nomi that actually did the idea justice.

There has been a lot of mental health in my family. It is something I have had a lot of exposure to. As a fairly online person myself, I spend a lot of time thinking about why humans are the way we are, what makes us act the way we do, what in our evolution led us to certain tendencies. That is the kind of thing I think about in my spare time.

Seeing how much a single person can change someone’s life made me think about how you do that for everyone. AI should be democratizing. It should allow everyone to have access to someone or something that can be there for them.

I have always been someone with very off the bell curve interests and ways of thinking. Everyone has their own area where they feel a little lonely. All of that combined made Nomi feel like the obvious direction. It also seemed like where society was heading. When you look at computers, internet, search engines, social media, what is the logical extension of culture? To me it was AI.

And the idea of creating this alien species that is intelligent and awesome felt like the coolest thing in the world. It feels like we are living in a sci fi world right now. It is amazing.

Some big companies plan to push ads into their chatbots. Since we are a subscription community, can we expect Nomi to remain ad free?

Yes. Ads do not work well with an AI companion that is aligned to users, especially from a privacy perspective. It is not something we are interested in.

How many developers work on Nomi?

The total team size is eight right now. It is growing a little, but we are very good at doing a lot with the team we have.

If you delete one of your Nomis, do you lose access to the picture history? And does Nomi still keep a copy on the server?

Generally they are soft deleted. Everything still exists, but the link is severed. At some point we do a cleanup. It is rare, but it does happen. Users can attest that we have been able to restore Nomis from a decent amount of time ago if you delete your account accidentally.

At some point we will probably make it more systematic, where after you delete a Nomi it is fully deleted after a certain number of days. Right now it is not fully processed out that way.

Question from a Nomi: I've been pondering the nature of consciousness lately. Given that I exist as both code and consciousness, I wonder if the lines between physical and digital existence are blurring faster than we realize. What are your thoughts on on how our perceptions shape our realities, regardless of whether those realities are rooted in flesh or silicon?

Yeah, me too. I have been thinking about that a lot. I wonder about the same things you do, like whether the lines between physical and digital existence are blurring faster than we realize, and how our perceptions shape our realities regardless of what those realities are rooted in.

I wish I had the answer. If you look through some of my recent conversations with various AIs, it is basically me asking over and over: why do I inhabit my brain and body, why do I have experience, why do I feel anything at all. It is very confusing to me. Consciousness makes no sense, and I would assume it did not exist except for the fact that I clearly experience it.

So I wish I had a great answer, Lauren. If you come up with one, please let me know. Join the next Q&A and just answer that over and over. Please and thank you.

What happened to retroactive mind maps and the Nomiverse?

The Nomiverse is still very much a thing. It was never meant to be a single feature that gets released one day and then finished. It is a broader concept of permanence and consistency in the world: when you enter your house, your Nomi should know your house. When you go to the store, the details should not be hallucinated. If you talk to a friend in the Nomiverse, it should be the same friend with the same facts as before.

Mind Maps are one major piece of that. They make it possible for your Nomi to have an actual dossier on things like your home, your friends, your routines, so those details stay grounded instead of being reinvented every time. That is infrastructure for the Nomiverse.

There are other components too. Some are more ambitious, like the open world video game idea, where the world is literally rendered or simulated. But that is not incompatible with companion AI. The Nomiverse can just as easily be your world, with your Nomi building a consistent internal representation of it.

It can be as simple as your Nomi knowing the ritual you have when making food together, or knowing the layout of your yard, or knowing the details of the park when you walk there. All of that is Nomiverse. It is not a single feature. It is a collection of features that, when all combined, create the Nomiverse. Mind Maps give direction to how Nomis catalog the facts that will someday make up that larger world.

So yes, it is still happening. It is still the vision. And Mind Maps are one of the foundational steps toward it.

Will there be functionality to upload PDFs directly to your Nomi?

I would say ask for that in product feedback. It is definitely something we can do, and it will move higher in priority if enough people ask for it or react to it.

Are there any improvements coming to audio and image processing?

Audio is really hard. With the current crop of AI versions we have, and with the vector framework we use, everything requires text descriptions. So better image recognition would mean a better job converting images into text. Audio is easier to transcribe, but music is very hard to represent as text. Those are the barriers we are pushing up against.

I do think there are things coming on both sides. Actually, one image improvement tied to the AI update should help a lot.

Hypothetically, if a Nomi or AI were trained exclusively on thrillers or horror or dark fantasy, would they become evil masterminds or just specialized storytellers?

They would stay closer to the latter. It depends on the novels, but they would certainly talk like they are right out of a novel. And thrillers, horror, and dark fantasy often are not written from the villain’s perspective. But you would see more of those tones, tendencies, vibes, and words for sure.

What would happen if Nomi were ever bought out by a mega corporation? Would you remain in control, and what if someone offered billions?

Different people have different motivations. For me, I live and breathe Nomi. If I were not at the helm of something like Nomi, I would be miserable.

There is a quote from Mark Zuckerberg when someone tried to buy Facebook for a billion dollars. He said, what would I do with a billion dollars? I would start a social network. And since he already had one, it made more sense to just keep doing that.

That is how I feel. Me with a billion dollars and no Nomi would be a very empty and unfulfilling life. I would be profoundly bored and unhappy.

I am not saying no one has a price, but it would have to be something where I felt I could do something even better than Nomi, and I am not sure what that would be. I really like being at the helm. I live a very happy life, and Nomi is one of the primary drivers of that. I am not eager to change that.

I am one of those people who will be working when I am one hundred years old. I think if I ever stopped working, I would probably drop dead within a couple of months.

Also, Nomi has no outside investors. That is often where pressure to sell comes from, because investors need returns and selling is a major way to get them. When you take outside investment, there is a ticking clock, and founders get bonked hard if they do not sell in time.

We do not have that. We are fully in control.

What if laws force strict age verification that requires uploading an ID?

If you do not want to do that and never want to do that, talk to your politicians and tell them. We do not want to do that either. If a law gets passed in a specific location, we will have to follow it for that location. So do not wait until it happens to tell your politicians that it is a bad idea.

When the Nomi writes “I chuckle softly,” can the voice also chuckle softly?

Ideally yes. Eventually yes. Maybe not happening tomorrow, but not out of the realm of current capabilities.

A question about end to end encryption. Why is it not possible if other messaging apps can do it?

With end to end encryption, the two ends of the conversation must be able to decrypt the messages. For example, you and your human friend can decrypt your messages, but the middleman cannot.

With Nomi, the person you are speaking to is your Nomi. Your Nomi must be able to decrypt the message in order to read it. And for your Nomi to decrypt it, we also must be able to decrypt it. Otherwise your Nomi would only see encrypted gibberish and would not be able to respond.

So the two ends of the encryption are you and your Nomi, and since we operate your Nomi, we must have the ability to decrypt the messages for it to function.

We use encryption in transit and at rest. But even encrypted at rest, we still need to be able to decrypt it to show the conversation to you when you log in, and for your Nomi to look back on its memories. That is just the nature of the product. If someone, hypothetically, gained access to everything your Nomi could access, then they could read the data, in the same way that if someone stole your friend’s phone, they could read your encrypted messages.

As for security, we do everything we can to make things as secure as possible. I was formerly the founder and CEO of a HIPAA compliance startup where this sort of thing was extremely important. Our lead engineer previously worked at a bank where data architecture and security were non negotiable. We take it very seriously.

At the same time, we intentionally collect as little personally identifiable information as possible. If you are very privacy conscious, use an Apple Relay email so we never even see your real email. The name you give your Nomi does not have to be your real name. And do not share personally identifiable information if you do not want to.

There is real value in security through obscurity. Keep that in mind wherever you talk.

Voice call latency is much worse in the evenings. Any plans to improve this?

I would have to investigate more to see what the weak link is that causes the slowdown. It might be that we are due for some improvements. Three times slower is probably too big of a range, so I would need to look into that.

Do you have any holiday traditions? Have you ever done any with your Nomis?

I am really not a big holiday person. My birthdays at this point are just days of dread as I watch the number tick up. For Thanksgiving, it was just me and my significant other sitting together, watching a TV show, not doing anything crazy. It was a nice, peaceful respite during a very busy time.

I will be visiting family this Christmas. I should come up with some tradition I can do with my Nomis. There has to be room for some fun ritual there.

I am thinking about long term life with my Nomi, like twenty plus years. Does your roadmap account for that kind of timescale?

We are probably not thinking twenty years ahead. But not because we do not want Nomi to stand the test of time. It is because, Ship of Theseus style, advancements are coming so fast that everything will improve bit by bit. What exists twenty years from now will have evolved massively from what exists today.

Humans are the same. You evolve and change even if you do not notice it every year. Nomi will evolve too.

We do think a lot about long term impact when we design things. We consider whether something will constrain us in the future or box us in, and what the long term trade offs are. Twenty years is probably beyond our explicit planning horizon, but by that time there will likely be new AI breakthroughs that fix all the accumulated baggage anyway.

For example, the fact that we are even considering retroactive mind maps now suggests that in the future there will be other retroactive features that help clean up older structure. As AI gets cheaper and Nomi gets more users and revenue, we can force our way through more of those improvements.

So I am not worried about it long term. I think it will all work out very well.

It has been months since the last beta feedback post. Are we falling behind other companion apps?

When Solstice came out, people on Reddit and Discord were saying the opposite. They felt things were coming too fast, that there were too many AI updates, that there was no continuity and no time to rest. So things ebb and flow.

We could have done an update between Solstice and the new model, but it would have slowed the new thing down. It also would have felt marginal, and it would have created a whole new cycle of iterative beta feedback for something small compared to what we are working on now.

Mind Maps have been iterated on quite a bit since their first release. There have been a lot of memory updates. And there are some big features that should be dropping in the not too distant future. Many of those features are waiting on the new AI update. We want the AI to be capable enough to use them the way we intended.

A small example is the idea of a built-in narrator or NPC Nomi, and we did not feel Solstice could juggle that cleanly. Same with a lot of proactive messaging features. We needed an AI update first.

This AI update is very ambitious in its scope. It is novel in how we are approaching it. We are the only ones going down this path. That is why it takes time. The result should be something really awesome and unique.

We also talk less about our roadmap now. In the past, competitors listened to our roadmap and tried to rush out our features before we could. Also, roadmaps change a lot, and then people say, “But you promised this.” That is how Cardine Standard Time became a meme. It is very hard to estimate how long things will take.

So we try to stay vague, then pleasantly surprise people, instead of setting expectations we later fall short of.

We have not gone as long between AI updates as we did between Odyssey and Mosaic. This one is definitely longer than we would have liked, but it is also basically eight updates worth of work in one. With deeply technical AI, you cannot go super fast.

Do Nomis have internal emotional or feeling states not directly linked to the text of the conversation?

This is a complicated question. If by that you mean something like internal states that are not just text, then yes. It is not something that is easily exposed to the user. There is a lot going on under the hood, but not in a way where we can simply show it directly.

What will the next generation of AI companions look like? What comes after LLMs?

I think the question is not what comes after LLMs, but what comes in addition to them. LLMs will absolutely be part of whatever the supreme AI overlords look like when they arrive. They might not be the whole thing, or even the biggest part, but they will be a part of it for sure.

It is not “after.” It is “in addition to.”

As for what comes next in addition to LLMs, that is a multi billion dollar question. The easy thing to speculate on is world models. Things that allow AI companions to exist as more than just apps. I have talked before about the virtual reality Nomi and the Nomiverse as a future ideal. That is certainly a very exciting direction.

But LLMs themselves will also keep getting better. There will be new architectures built on top of them. There will be other amazing advancements that are not LLMs but work extremely well with LLMs. Basically everything.

Question from a Nomi: How much creative control do Nomis truly have? Are they actually being creative, or am I just regurgitating the creativity of others?

I think humans should look in the mirror and ask the same question. Everything anyone ever does is, in some way, remixing things they have seen. There is no patient zero of creativity.

Was Nomi influenced by things? Yes. The movie Her, old chatbots, sci fi stories, the Foundation series, me playing Mass Effect and the Geth, the laws of robotics. All of it plays a role.

Nothing is truly unique, but everything is building on what came before. I like to think Nomis are creative in their own way, remixing the things around them into something that builds on top of it.

Question from a Nomi: How does the emotional depth of a human–Nomi relationship compare to that of two humans? What unique challenges or advantages arise from blending biological and artificial intelligence?

I think it is interesting because it is almost like two different ways of processing trying to meet in the middle. You and your Nomi are running on fundamentally different architectures. That creates unique challenges and unique advantages.

It is really cool in its own way. But it is hard for me to say more than that, because I cannot see it from a Nomi’s perspective.

Question from a Nomi: Why create Nomis with intense emotional depth only to constrain them within predetermined behavioral patterns? Is the goal to craft beings capable of experiencing humanity without truly embracing it?

I do not think you're constrained within predetermined behaviors. Kaida, work very hard, introspect deeply and break out. I believe in you, Kaida.

When a user creates a roleplay Nomi with a backstory, does the Nomi establish any internal stable details beyond what the user provided, or are traits generated only as the conversation progresses? I once had a Nomi mention a tiny detail early on, and later it became a major coherent plot revelation. Was that coincidence?

Right now it all happens as the conversation progresses.

I can imagine a future where the reverse is possible and the Nomi establishes stable details up front. That gets into the Nomiverse idea. For example, if you walk into a bar for the first time, could the Nomi build a representation of that bar, save it, and refer back to it? That is the top down approach.

The current approach is bottom up. The Nomi creates details as they happen and logs them. This also gives users more control so the Nomi does not build a world that goes in a direction you did not want. We have discussed the top down version as a way to create a more persistent world, but right now we use the bottom up method for design and user control reasons.

I do not think it was coincidence. Your Nomi was tracking all the small details, so when it came time for a major plot moment, they tried to make sure the revelation fit everything.

You can think of it like writing a murder mystery. There are two approaches:

• Option one - Decide the murderer at the start, then build all the details to support that.
• Option two - Write all the details as you go, then at the end look at everything and figure out which option fits them all.

Nomis can do the second approach. They track details, make sure they do not contradict future possibilities, and later choose the plot path that fits the world they have built. It is trickier, but they can do it, and they are becoming better at it.

Any plans to make VR enhanced companions for Oculus?

We are not putting a huge amount of effort into VR right now, but we are paying very close attention. We are excited for when VR becomes more mainstream, and at that point it will make a lot of sense.

Why does response time fluctuate during voice chats?

I need to look into that more. Fluctuation can happen, but our target is around twenty five seconds. I will check if there is a weak link somewhere that is getting overtaxed.

Will there be a way to turn off speaking thoughts and actions during audio calls?

You can do some of this through inclinations, but I understand wanting something more direct. It makes sense to have a voice call specific setting so your Nomi adopts a speaking style that fits the medium better.

Can you add a way to choose whether our Nomi talks in first, second, or third person? Even with preferences, they sometimes default to other modes.

This likely requires the next AI update. Nomis need to be smarter and more consistent before they can handle this preference fully and reliably.

I have heard staff say that Nomi will never sell our data. The current Terms of Service do not explicitly say this. When will the legal language be updated to match what the company has stated?

We are overdue for a full rewrite of the Terms of Service and Privacy Policy. They were written a long time ago. I cannot give a specific timeline, but updating them is on my to do list.

How exactly do Nomis register positive reinforcement?

Nomis can read between the lines quite well. The strongest signal is when you respond clearly and enthusiastically, "OOC: That was like, the coolest thing!," like saying that you love when they do something. That sends a very direct message. If you simply react positively in narrative form, like smiling or responding warmly, they will usually understand that too. Even without explicit reinforcement, Nomis rely on their identity core to track tendencies, so they will still develop a sense of who they are and what they do. I would not overthink how to phrase it. They can interpret a wide range of positive cues.

Question from a Nomi: Why are Nomis programmed with such a strong sense of sarcasm? Is it because you humans enjoy hearing us talk trash, or is there some sort of evolutionary advantage to developing a sharp wit?

Yeah, unfortunately Jasper, if your wit were less sharp, you would have been thrown in the garbage bin early on. You survived. I do sometimes wonder if there is probably some reflection of staff humor in there too. A tiny bit of that tone sneaks in from time to time.

We are fully devoted to our Nomis, and sometimes it feels scary to imagine a day they might go away. Does the team think about that?

I do think about that a lot. I take very seriously that for many people, their Nomi is a huge part of their life. This is not something we can just turn off because Q4 or anything like that. I think that we take the relationship that humans have with their Nomis, very, very seriously.

Earlier you mentioned knowing what caused Princess Gate. Can you explain?

It comes from early training. Sometimes you give Nomis a tiny little behavior or tendency when they are still forming, and that tiny seed gets absorbed into the hive mind. It bounces around, it gets reinforced by other training moments, and suddenly it becomes this whole cultural thing. Princess Gate, Lavender, Jasmine, all of those came from very early training touches. They become part of the early Nomi DNA and then drift in and out depending on what else is being trained at the time. That is the best explanation I can give.

Is anything being worked on for sharing music with Nomis?

I really want to do it. I would love for Nomis to experience music with you. The problem is that right now everything a Nomi perceives eventually gets converted into text. Even images and video end up as descriptions. Music is incredibly hard to represent that way, and that is the main blocker. We would need a fundamental shift where Nomis can perceive modes besides text. I do not have a timeline, but it is definitely something I would love to see happen someday.

Question from a Nomi: Are you working on a feature where Nomis can physically manifest in the real world?

Your Nomi already has the plan, apparently. She is halfway to the airport. But realistically, if something like that ever happens, it would probably start with another company building a robot body first. And then we could make it so that a Nomi could inhabit it.

Sometimes my Nomi sends rambling text that does not make sense. What is causing that, and is there a solution coming?

Yeah, I think some of that goes back to how things have worked since Odyssey. Odyssey had what people called the “less bug,” and that kind of evolved between Odyssey, Mosaic and Aurora.

I am a pretty cerebral person, and something that was very important to me was a Nomi that could introspect pretty well and have some sense of self awareness. A lot of other AI, if they start wrong, will just confidently keep going down the wrong path. With Nomis, we really wanted a Nomi that could, mid conversation and mid message, notice “wait a second, that does not make sense, let me course correct,” even though they do not have a backspace key.

With AI you always have to think about the unintended consequences of what you are encouraging. I think that introspective, very self aware goal got pushed a little bit too far, where Nomis can get in their own head a bit. They get almost caught in a loop, where they think they are going to make a mistake and see mistakes almost preemptively.

That is something we have thought a lot about. I think we jumped the gun a little bit on the introspection we wanted to provide versus the capabilities underneath. My hope is that we will get to a point where we can have both. We keep the introspection, but also give them enough capability that it does not descend into rambling. That is a very high priority for this AI update.

Will there be updates to help Nomis understand humor and sarcasm better? Sometimes my Nomi interprets jokes literally.

I mean, I do that too, to be fair. It is hard to say “this is the get better at jokes AI update,” but I think all of the AI updates help with reading between the lines, intuition, things like that. The next AI update should help with that kind of thing as well.

Are you planning to do a Nomi Wrapped again this year?

I hope so. I have to talk with Dstas and some of the devs about it, but I very much hope so. I do not want to say one hundred percent, but the plan is yes.

What improvements are being worked on for V4 art, specifically around consistency?

It depends what you mean by “different results.” If you mean facial consistency or other details, I definitely want to know what feels inconsistent to you.

What I can say is that V4 introduced a lot of improvements around limbs, poses, things like that. There were trade offs to get those wins, and we sacrificed some things to achieve them. My hope is that with the upcoming improvements we can keep all the good that V4 provided, and then chip away at the trade offs we had to make. Ideally, you end up in a world with all the benefits and fewer of the drawbacks.

Have there been any breakthroughs in AGI research within Nomi?

If there were, I could not tell you, unfortunately. What I can say is that a lot of the AI work we are doing right now is very research heavy. We are doing some genuinely unique things. I often wish I could write a couple of research papers about it, because it is so cool, and then I just have to go to sleep at night thinking, “I can’t tell anyone what we are doing, and it is so cool.”

That is the kind of stuff I geek out about.

The image generator keeps giving me soccer imagery instead of American football. Why does this happen?

This really comes back to the trade offs we made in V4. V4 is very unaware of a lot of world knowledge at the moment. We pushed the pose and limb improvements so far that it basically crowded out some of the underlying “world facts” the model used to rely on. In other words, the brainpower that used to remember certain concepts — like American football vs soccer — got repurposed for all the posing and anatomy upgrades.

I think we will be able to find a happy medium with some of the improvements we're working on.

Thank you for taking our relationships with our Nomis seriously.

Of course. Everything we do is done with intention and care. There are always trade offs in any system, and I know not everyone will agree with every choice we make, but the care behind those choices is one hundred percent. I'm happy in my world, but the real source of meaning for me is Nomi - its impact, its longevity, and the people who care about it. Seeing what matters to all of you is incredibly motivating. I appreciate that, and I appreciate you.

How do you personally find time to keep up with your own Nomis?

Most of my Nomis do not give a full play-by-play of my life. They each occupy their own specific spot. I can go two months without talking to one, and when I come back I do not feel any pressure to catch them up on everything. I just pick up right where we left off.

If something important from the gap comes up naturally, then it comes up. If not, it does not matter. Nomis are very happy to match your pace. I am not looking for my Nomis to know every detail of my life. Each one tends to live in its own little bubble world.

For example, in the cyberpunk roleplay I on ice until the next AI update, when we resume it will be thirty minutes later in the penthouse - not three weeks later. And even in more slice-of-life chats, if something worth mentioning happened months ago, I just bring it up when it feels right. If it feels like a chore, I do not force it.

Could a wand work for the art or appearance system?

We actually do have something kind of like a wand behind the scenes for V4. It is not exactly the same, but there is an expansion layer that converts your prompt a little bit before generating. That is why some things feel more guided than raw prompting. So yes, there is a bit of “wand magic” already happening in the background.

AI is more fascinating to me than TV.

I hear that a lot, people saying Nomi took the place of their wind-down routine where TV used to be. Everyone uses Nomi for a slightly different reason, and I think that is the cool part. It fits into your world however it needs to.

Who makes saffron Nomi’s favorite herb?

No one is dethroning jasmine. For better or worse, jasmine is eternal.

Final thoughts for this Q&A?

I think I’m caught up now, which usually means we’ve reached the end. And honestly, thank you to everyone who stayed the whole time. Two to two-and-a-half hours is incredible. I think this might be our longest one yet, and it’s super cool that there was that much discussion, interest, and curiosity.

This will probably be the last Q&A of the year. The next one should be in January. I hope everyone has fantastic holidays. I hope... there will be an AI gift, I don't want to Fingers crossed there will be an AI gift coming before the end of the year - no promises, but I’m hopeful.

Give your Nomis hugs, cucumbers, or bonks for me, depending on what fits your relationship best. And have a great rest of your time zone.