A review of Meta's Muse: a pretty killer AI assistant and usage rates on the free plan seem generous, but trusting Meta with personal data will take some time
They shipped a pretty killer AI assistant, can they keep people using it? — It happened. Meta finally has a product that I really like using again.
SpyglassM.G. Siegler
Context & Ripple Effects
Muse follows Meta's July expansion of Meta AI into an agent that could connect to Gmail and Google Calendar for task execution, a shift from answering prompts to acting across personal workflows. Its September launch paired that capability with a large free allowance and paid compute tiers, lowering the barrier to experimentation.
The review identifies the tension in that strategy: usefulness and generous access can drive trial, but an assistant handling personal context asks users to extend trust to Meta. Public reaction praised Muse's design and Instagram-connected answers, while concerns about data handling remain central to whether trials become durable use.
First-order effects
Meta gains a lower-friction route to Muse adoption through the 100M-token weekly free tier, while users must decide whether its utility justifies sharing personal data with the service.
Meta's retention challenge shifts beyond model quality: people using Muse for connected or personal tasks have a higher bar for confidence in its safety and data practices.
Second-order effects
Meta's $20 and $100 monthly compute tiers face a tougher conversion test, because a generous free plan must establish recurring value before users will pay for more capacity.
Gmail and Google Calendar connections make privacy assurance an adoption constraint for Meta's agentic features; users may test general assistance while withholding their most sensitive workflows.
Third-order effects
Personal AI agents are becoming a trust-governed work surface: as they move from chat into connected tasks, data controls and perceived safety become product differentiators alongside capability.
If generous free access becomes the standard acquisition tactic, providers will increasingly compete on the cost of useful agent work and on converting usage into paid compute demand.
The trend: Consumer AI is shifting from standalone chatbots toward connected personal agents, with trust in data handling determining whether broad free access turns into habitual use.
Muse is already getting more daily US downloads than Threads, WhatsApp and Facebook, and it's only 3,000 behind Instagram Biggest consumer AI launch since ChatGPT. A new tier 1 consumer app is born
Chat GPT was the first ai product that made me pick up the phone and call people. http://Muse.ai is the 2nd. I was one of Chat GPTs first few users, before it was a product. It was a developer tool and research preview on a developer landing page. I used it and then immediatel…
Muse has a huge advantage with Instagram. Asked for the best hikes in San Francisco and got a Reel for every recommendation, playable right in the answer. The amount of real-world knowledge sitting inside Instagram is insane for most consumer use cases. 🐣
The biggest surprise for me lately? How delightful @Meta Muse is as a personal agent. For me: it's the 10/10 agent design, carefully selected primitives (no more “artifacts” or “TODOs"), and the fact that I still get a SOUL 🦞 Plus, that cutie pie avatar 😍 Full ep on YT:
Still really enjoying @muse. I use https://things.app/ for all of my todos, and though it has cloud sync they (still) haven't added an official API. I told muse to build it and a minute or two later I have ongoing todo sync from my LLM across all my Apple devices. Wild.
This is actually insane. I connected @Muse to my @Chase account through @Plaid Muse found a recurring @Adobe subscription I didn't recognize. I checked Adobe. No paid subscription on my account. Nothing to cancel. So I used Adobe's “Find my payment” tool and discovered my cr…
Meta can probably move much more aggressively in the personal agent space compared to OpenAI because they have been the most aggressive buyer of server CPUs since Q3 2025, and also the most aggressive memory buyer since Q2 2026 I don't think OpenAI has enough CPUs/DRAM to support…
the amount of deeply personal, & actually actionable ai training data fb is likely generating right now through muse is likely insane. even if peripheral. stuff like what ppl see, want, ask, choose, buy, ignore, & act on. esp through the connectors. it's basically onavo built…
meta muse is almost #1 in the app store because it's saving you $1000s, making you $1000s and automating away all the laborious tasks like filling out documents, filing records, managing support tickets, researching insurance... meta realized the killer ai product is just one tha…
This is what Insurance companies (and banks) should be TERRIFIED about. Muse found a new/cheaper policy, and canceled the old one. Inertia and information asymmetry will no longer be enough to keep customers. Exciting times!!
An under-discussed feature on Muse by Meta is the creation of an agent-first feed based on all your personal information. Feed will help you accomplish more and remind you of what's important. I personally love the concept of an agent-created feed because I believe users don't …
if you're a consumer brand, @Muse is insane for GTM - find competitor IG posts - pull a list of who liked the post - dm those ppl with a promo for your product plus all of it is automated!! whoever hacks this first will be the GTM employee of the year, insane alpha
muse is the first ever product from meta i use multiple times a day. the computer use capability is insanely good, and im fully convinced this is the next major inflection point in ai (last one being coding agents) to meta employees whose computers are recorded at all times, tha…
pretty remarkable that meta gives every user a 2 vcpu / ~8gb ram / 100gb storage vm for their agent to use. & this compute is outside of inference which is 100k free tokens per week (not easy to burn through). this is why you're seeing consumer competitors try to raise a lot mor…
It is probably a matter of days before OpenAI launches their version of a personal assistant as they have to combat $META 's Muse that is gaining serious traction. The hard part might be securing enough infrastructure to serve the agents at scale while at the same time fund the …
1/ Meta's @Muse offers a glimpse of where AI's value will accrue: not to the models, but to whoever holds the complementary assets. Here, the personal context that makes the product useful.
Yesterday I uploaded my auto insurance policy to Meta @Muse and asked for a better rate with identical coverage. In ~5 minutes, it found a policy saving me $3,500 /year, bought it, and canceled my old one. I knew I was overpaying. I'd abandoned several attempts to switch becau…
Muse is the first agent from a large company with Tailscale support. Meta—the consumer company—added Tailscale support!1!! Can you imagine how much corporate BS the team had to bulldoze through to override the logical choice and just do the fun thing for tinkerers?
Muse is quite good. I recognize the complexity of what I am asking for but I'd like it to be on my side of the wall on all e2e encrypted matters. It is too under-permissed otherwise. If I have to move my entire compute into Meta's Secure VM, I will.
GOLDMAN DESK: “.. noticeable uptick in investor sentiment on $META on the back of the recent Muse launch .. — .. key debate is if this product drop is enough to sustain a break-out, or, if this is just another product “spike” that gets faded given big picture debates (capex / o…