ROBLOX script still allows messages through

Question

I am making a ROBLOX script to block messages which are toxic from getting to other people. For some reason, it isn't working and still allows the toxic messages to be seen by others. Any ideas on how to fix it?

I have tried a super basic filter, which just removes words it finds to be “bad” from a table, and that surprisingly works.

ONLY SCRIPT:

local TextChatService = game:GetService("TextChatService")
local Channels        = TextChatService:WaitForChild("TextChannels")
local HttpService     = game:GetService("HttpService")

local TextClassifier  = "https://api.example.com" -- not actual thing, actual api returns json data
local MaxScore        = 0.7

local function isMessageToxic(message: string, speaker: TextSource): boolean
    if message == "" then return false end
    
    pcall(function()
        response = HttpService:PostAsync(
            TextClassifier,
            HttpService:JSONEncode({
                text = message
            }),
            Enum.HttpContentType.ApplicationJson
        )
        response = HttpService:JSONDecode(response).response
    end)
    
    if response.Error or not response then
        print("There has been an error.") -- no body or etc
        return false
    end
    
    if response[1].score > (MaxScore or 0.7) then
        print("Message has been flagged.")
        return true
    else
        print("Message has passed.")
        return false
    end
end

local function channelAdded(channel: TextChannel)
    if not channel:IsA("TextChannel") then return end
    channel.ShouldDeliverCallback = function(message, textSource)
        return not isMessageToxic(message.Text, textSource)
    end
end

for _, channel in pairs(Channels:GetChildren()) do
    channelAdded(channel)
end
Channels.ChildAdded:Connect(channelAdded)

What is sent to the API:

{
    "text": "i hate you”
}

What it responds with:

{
    "response": [{
        "label": "NEGATIVE",
        "score": 0.9993651509284973
    }, {
        "label": "POSITIVE",
        "score": 0.0006348910392262042
    }]
}

Image of it not working:

What is printed:

17:19:59 -- Message has been flagged.
17:19:59 -- Message has been flagged.