Something weird has been happening with my phone for years. I thought I’d fixed the issue, but it always comes back. Years ago, I bought a SIM-only plan through Visible, Verizon’s cheap sibling. $25 a month gets me unlimited calls, texts and a decent amount of data… and a lot of spam calls and messages looking for someone called Andrew.
Given Visible is all-digital with no number to call, it’s a pain to wrestle with a customer service chatbot to fix it. I’ve tried before, and it said the number is in my name, but I’m not convinced. I still receive multiple messages per month from people trying to reach Andrew.
These messages almost always have something to do with a home trades project. Weird, I know.
Screenshot by Amanda Smith/CNET
It’s one of those life admin tasks that just never gets done because I can easily ignore these messages – making it the perfect job for a personal AI agent like Meta Muse. I want to make sure my name is formally listed as the number owner, as I don’t want people thinking I’m this elusive Andrew.
I tried using Meta Muse as my personal shopper, which was fun, but I ran into an issue at checkout. Let’s see how it can handle an annoying task like this, without a number to call. One AI agent versus another.
Setting up Meta Muse with the task
I opened the Muse app and started a new conversation, instead of picking up on my previous chat with the AI agent. Keeping tasks and chats separate prevents context overload and helps Muse stay focused on the specific objective at hand. Just hit the new side chat button.
Here’s how I instructed it: “I am a Visible customer on the $25 per month plan. I often get calls and texts looking for someone called Andrew. I think he was the previous owner of my number. It’s not a coincidence, because every time someone reaches out, they always ask for Andrew. I’ve contacted Visible before, and it said I own the number, but can you check I’m formally listed as the number owner? Check government websites, etc.”
I also uploaded an example of a text I got.
This was Muse’s response:
Meta/Screenshot by CNET
I told Muse to talk to the Visible chatbot, search my number online and find anywhere that references Andrew. Within seconds, it found a link where my number showed up on a data broker page.
Meta/Screenshot by CNET
But Muse couldn’t get through, as Visible blocks bot visits. Muse suggested changing my number (no) and making some changes on my phone to minimize unknown senders and silence unknown callers. This wasn’t super helpful either, as I often receive work calls from numbers I haven’t saved.
I told Muse I didn’t want to change my number. It advised me against claiming the number, as that can encourage spam calls directed at me instead of Andrew.
It said the Andrew data will go stale at some point.
Meta/Screenshot by CNET
I had to take over the browser for Muse to chat with Visible. That was taking forever, so Muse did a workaround via the guest option. This took a good 10 minutes, which I spent cleaning the kitchen. But it had no luck.
“Quick honest update,” Muse told me, “I got into Visible’s guest chat, but their send button wouldn’t submit my message, and then their help pages temporarily blocked me.”
It advised me to paste the following chat into my Visible app: “My number is recycled — I get monthly spam calls and texts for ‘Andrew,’ the previous owner. I don’t want to change my number. What are my options? Do you offer spam/scam protection, and how do I turn it on? Can I report the number as recycled?”
Of course, such advice could have been delivered to me by a regular AI chatbot like ChatGPT or Gemini. The advantage of using an AI agent like Muse is that it’s supposed to do all of that for me.
But with a lot of these websites blocking AI agents, it makes autonomous task completion a challenge. I attempted to take over the browser again, but got a 403 forbidden error.
Meta/Screenshot by CNET
Muse told me the 403 error is a dead end and that my taking over the browser didn’t change the network address, which was annoying. Muse said it’s the bot protection that’s the problem, not Muse. It drafted a message to send to Visible chat manually, via a live agent.
Screenshot by Amanda Smith/CNET
While I chatted with Visible, Muse gave me a link that showed Andrew and my number. What’s interesting is that the last reported date was when I owned the number.
When I ran a quick manual search on my phone number, the top two websites referenced Andrew. I asked Muse how to update these, and it gave me good advice, like requesting a suppression or removal of records from WhitePages and FastBackgroundCheck, but no instant resolution.
Muse said this would clean the public record, but the messages are coming from old marketing lists, so it said I should keep reporting those as junk. Brokers re-scrape, so it should resolve in time.
I tried it via WhitePages, but quickly hit a roadblock. It wouldn’t let me click Next.
I couldn’t follow the second instruction as I don’t have Andrew’s email address. It advised me to use my email (no thanks) or to call the number.
In another browser, I got escalated to Visible’s “advanced” support. Its instruction was to keep the browser open for 4 to 24 hours.
At this point, I gave up.
The verdict on Meta Muse for this task
I wouldn’t say this was Muse’s fault specifically. It seems most websites have mechanisms in place to block any bots. Most AI agents would struggle to complete this autonomously. For me, in this situation, Muse still feels like a regular chatbot, giving guidance for me to follow.
While I better understand the issue — spammers using outdated data — it still didn’t fix the problem. I’m hoping Visible pulls through with a solution and, if so, this was still time well spent. Without testing Muse to see if it could do it for me, I wouldn’t have prioritized this task.
So, that’s a win, I guess.
See full bio