Amazon says it is involving the most recent in man-made reasoning (computer based intelligence) to get serious about counterfeit audits and recognize remarks that aren’t certified.
Fake reviews “brokers” have been a major issue for the tech giant’s online shopping site.
In order to assist it in identifying the fraudulent behavior, Amazon has invested in machine learning models that analyze thousands of data points.
But Which is a UK consumer group? says that there is still “nowhere near enough” action.
Fake review brokers buy, sell, and host fake reviews on third-party platforms like encrypted messaging services and social media.
Counterfeit surveys can influence clients to pursue buying choices, for instance over which PC or kids’ toy to purchase, in light of what they accept is veritable criticism from different customers, when as a general rule somebody has been paid to compose a shining audit to help a dealer’s evaluations, or to sabotage an opponent firm.
They are generally difficult to recognize, albeit conventional data, or an exceptionally high level of five star surveys can be a part with.
Over 23,000 social media groups with over 46 million members and followers facilitated fake reviews, according to Amazon in 2022.
Amazon has been using AI to combat fake reviews for a while, but the company claims that investing in more “sophisticated tools” should make its platform safer for sellers and customers.
According to the company, its fraud-detecting AI was able to determine the likelihood that a review was fake by looking at a variety of factors. That can incorporate the creator’s relationship with other web-based accounts, their sign-in movement, survey history, and any strange way of behaving.
Dharmesh Mehta, the head of Amazon’s customer trust team, told the BBC, “We use machine learning to look for suspicious accounts, to track the relationships between a purchasing account that’s leaving a review and someone selling that product.”
He stated, “We can stop those fake reviews before a customer ever encounters them through a combination of both important vetting and really advanced machine learning and artificial intelligence – that’s looking at different signals or behaviors.”
Not Enough
Too little Harry Kind on Which? said that, according to some estimates, one in seven online consumer reviews are fake in the UK.
“Amazon has been utilizing a variety of technologies to combat fake reviews, and it appears that this strategy is working.
He stated, “But as far as we’re concerned, it’s still nowhere near enough to solve this enormous problem.”
The consumer group asserted that consumers were more than twice as likely to select low-quality products due to fake reviews.
Because of the new strategies it had created, Amazon said it had obstructed north of 200 million thought counterfeit audits last year and would “keep on building refined instruments that safeguard clients”.
However, in order to increase the strategy’s efficacy, the retail platform calls for increased collaboration between the private sector, consumer advocacy organizations, and governments.
The Competition and Markets Authority (CMA)’s legal authority in this area is expected to be strengthened by the Digital Markets, Competition, and Consumers Bill that is currently in the UK parliament.
Which? While acknowledging Amazon’s call for a more collaborative approach, the company urged the UK government to “explicitly make the buying, selling, and hosting of fake reviews subject to criminal enforcement” in its proposed legislation.
On Facebook, it said that fake review “factories” still had easy access to trade reviews for Amazon and other sites.
The CMA stated that it had already taken significant action against those who trade fake and misleading reviews.
The CMA’s spokesperson stated: Our examinations connecting with counterfeit audits – including the case into Amazon – are progressing and further updates will come in the not so distant future.”
Legal Action
Recently, Amazon filed a lawsuit in the United Kingdom against the owners of NiceRebate.com, a phony review broker aimed at British customers.
Different sites show to similar administrators were likewise closed down, with synchronous legitimate move made against them in the US.
Mr. Mehta stated, “We are aggressively fighting review brokers.”
He stated that Amazon had taken legal action against 94 of these “bad actors,” which included con artists in the United States, China, and Europe.
Amazon fake reviews checker reddit
Here are some of the most popular Amazon fake reviews checkers that have been mentioned on Reddit:
- Fakespot is a website that analyzes Amazon product reviews and filters out reviews that its algorithm detects may be unnatural. It gives each product a grade from A to F, with A being the most trustworthy and F being the least trustworthy.
- ReviewMeta is another website that analyzes Amazon product reviews. It looks for patterns in reviews that may indicate they are fake, such as all of the reviews being written in a similar style or all of the reviews being posted around the same time. ReviewMeta also gives each product a grade from A to F, with A being the most trustworthy and F being the least trustworthy.
- Crowdsourced Reviews is a website that allows users to vote on whether they believe a review is genuine or fake. The website then calculates an overall trust score for each product.
- CheckBack is a browser extension that allows users to see how many reviews for a product have been deleted. This can be a sign that there are a lot of fake reviews for the product.
It is important to note that no review checker is perfect. Some fake reviews may slip through the cracks, and some genuine reviews may be flagged as fake. However, these tools can be helpful in identifying products with a high number of fake reviews.
Here are some tips for spotting fake Amazon reviews:
- Look for reviews that are all five stars or all one star. This is a red flag, as most products will have a mix of positive and negative reviews.
- Check for reviews that are written in a similar style. Fake reviews are often written by the same person or group of people, and they may have a similar writing style.
- Look for reviews that mention specific keywords or phrases that are often used in fake reviews. These keywords and phrases may include “free product,” “great deal,” or “I was not paid to write this review.”
- Beware of reviews that are posted shortly after a product is released. It can take time for genuine customers to receive and use a product, so it’s suspicious if a product has a lot of five-star reviews shortly after it’s released.
By following these tips, you can help to avoid buying products with fake reviews.
One thought on “Amazon cracks down on fake reviews with AI”