In the age of social media and AI algorithms have the power to decide everything, our playlists to the videos we watch, to the news we consume and the shopping specials that are offered to us, and those that are refused to us.
Despite all the good that technologies and machine learning algorithms do to improve and personalize the online experience for all of us, they also represent a major online security threatswith real negative implications for the health and well-being of all Internet users.
Used carelessly and in the absence of safeguards and safety programs, algorithms often harm users they are meant to serve. Predictive algorithms can trick platforms into continually delivering content related to thinness and eating disorders, suggest you watch more extremist content, or amplify polarized beliefs.
The effects of the bias of the machine-learning are just as scary. Bias negatively affects left and right, liberals and conservatives no group is spared and the effects of extremist, polarizing and unhealthy content online on the human behavior in the real world have been well documented.
Ban the algorithmic recommendations like those made by TikTok or Instagram for children and make it an “opt-in” choice for adults. We still understand too little about how algorithms push harmful content and the damage it causes. A more cautious approach, especially for vulnerable groups, should be a security priority.
Integrate systemic and routine testing into the process of development. Independently certify new algorithmic recommender systems prior to launch to confirm match between intent and the positive impact. Scientific proof that a technical system does what it was designed to do without negative side effects for users is a standard safeguard for consumer protection in almost every industry.
Be transparent about the state of the systems and risks to users. Recommender systems have a huge impact on a very large number of users. As such, it is imperative that the general public understands the behavior of these systems. Measurements, tests, compromises, inputs and observed impact must be made available to users, regulators and academics for analysis.
The impact is too great and the risk of harm too severe to remain an unregulated space. The same way we test new drugs, cars, or nuclear power plants, we need sensible regulation that protects the general public from harm.
artificial intelligence is most important technology of our generation. It is not the responsibility of any single organization or entity to guide responsible development and provide control and measurement. It will take the combined collaboration of private companies, public entities, advocates and policy makers.
We wish to give thanks to the writer of this short article for this amazing web content
Should recommendation algorithms be stopped?
You can find our social media profiles and other pages related to it.https://www.ai-magazine.com/related-pages/