( January 28, 2022; IEEE Spectrum )
Description: "One machine-learning-based system was created by a group at Georgia Tech is a 'virtual assistant' that screens callers with a few questions, like 'whom do you want to speak to?' By judging the answers, the AI can guess as to whether a call is legit."
Ahamad describes the system as a “virtual assistant” that screens callers with a few questions, like “whom do you want to speak to?” or “what is the weather like where you are calling from?” By judging the answers—or details like whether a caller interrupts the questions—the natural language processing system can make an educated guess as to whether a call is legitimate (or wanted).
It’s entirely possible that robocallers might deploy natural language processing systems of their own. But, Ahamad says, “It would require lots and lots of resources, if you’re going to make millions of calls, to be doing that for each and every target that you have.”
SPAM is not just a problem for Steem.
In the past, US telephone carriers were required to deliver all calls. That changed a few years ago, so the telcos were permitted to drop calls that didn't originate from an actual phone number, and with the 2021 STIR/SHAKEN mandates, phone companies are now required to block some SPAM calls (smaller companies get until July/2022). Additionally, the emergence and continuous improvement of cell phone spam blockers - as above - tilts the balance even further against the SPAM callers.
What can abuse fighters on Steem learn from SPAM fighting in other domains?
Read the rest from IEEE Spectrum: AI Can Sniff Out Robocallers
Check the #penny4thoughts tag to find other active conversations.
###### This markdown/html was auto-generated by the java Steem Links Creator prototype application.
Hello @remlaps-lite, I think that technology helps us detect not only SPAM, steemit can use many of these tools not only to prevent SPAM, but also the immense monopoly that exists in the panel, conclusion, they should work more and not stay alone with the POWER UP that most of us do
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit
To fight against SPAM, our Steemit platform can implement CAPTCHA or some other challenge–response test every time a user is publishing a new post or comment.
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit
True, but people can always post to the blockchain by other mechanisms than the Steemit web site. I don't think there's a way to require a captcha from all possible posting sources.
But now I'm wondering about some sort of captcha-based verification service. Maye a service could be set up where people could voluntarily complete a captcha periodically and then curators could use that information to guide their voting. Here's something of the kind for Ethereum.
To be honest though, as a user I'm not a big fan of captcha, so I'm not sure if I would participate in something like that or not.
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit
Spam in steemit is like fish in the sea, it is almost impossible to kill because even if you kill most of them, they will always reproduce en masse.
I would like the spam that comes to me by email to end, but this is another problem that I think google cannot solve.
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit