These are serious claims and seem to be a provocation.
As for the technical part. There is no question of any dynamic and static analysis of speech. Confirmation of a false request from the user
is also described in the algorithm for assigning the level of risk. Read carefully how the threat level is set.
Reputational base formation
A novel concept of weighting certain attributes for the determining the Risk Level of
potential software packages. This Risk Level is evaluated by the LevelNet system
automatically based on the value of a set of factors listed below in descending order of
importance:
● antivirus solutions reactions from end-point users.
● level of trust that its publisher has
● prevalence within the network
● the number of cases when this software was added to the list of exceptions
end-users
The number of antivirus solutions can be obtained both from users' devices
and from the internal servers of the LevelNet network. The process of monitoring
previously unknown software.

Also read about the database of trusted publishers, you most likely missed it or just did not know it.

If the object (the new hash) does not satisfy the conditions, then it is considered dangerous, thus all manipulations with the cryptor and morphing are reduced to zero due to the fact that the object
previously did not meet with users and there are no trusted publishers in the database.
The algorithm is specified in the answer number 1. In addition to this algorithm, the number of requests from the user is checked. In reality, if the user
sends many requests, it is blocked due to the fact that this is not a real behavior. Because sending a lot of requests from the user in reality does not happen.
2.1 We do not check each file, we use the algorithm specified in item 1. The source of information is user devices and the database of trusted publishers and their files.
We do not use dynamic anti-virus scanning on our side.
3. It's not the number of devices, but the speed of delivery of verified information. The data transfer rate is very high today. There's really no verification on the video
on the reliability of the request received from the device on which an antivirus is installed, which is able to cope with the threat. Validation test
A request from a user for an identified threat does not take much time. We do not analyze the set of anti-virus engines in our
in the cloud, but we operate the information with information obtained from other users and sources of re-checking the information.
Verification of the request is not based on either dynamic or static analysis, which you specified in paragraphs 1-8.

On the legal side:
- All results of anti-virus software on the user's device are the property of the user under license agreements.
Number 9 Are you sure of this? If so, I ask you to provide evidence that we have not undergone any independent technical audit.