Google just revealed in its yearly I/O developer’s conference how its biggest mission was to enhance the quality of search for users through the help of AI technology. But it clearly seems to be falling short of its promises after the launch of its controversial AI Overviews feature.
Not only is the offering giving out inaccurate replies at the speed of light but there’s zero accountability from the company with no plans of getting things fixed. Did we mention ads are being included too and it appears that it’s giving rise to huge frustrations across the board?
Turning to the search engine giant to get a great response is just mind-blowing for many to comprehend because Google is doing the exact opposite of what it’s supposed to do.
We agree the feature isn’t toggled to interfere in every single reply but the fact that it goes through the web to produce AI-based replies has people talking. So far, we have seen one query about pizza glue make the rounds and the reply generated was taken from an ancient thread posted across Reddit nearly 10 years back. Mind you, that was a joke on the app but AI Overviews really did take it seriously.
This happens to be one of the long list of mistakes popping up through the latest feature online which Google published in May. But it’s the start of a huge list of wrong answers that has people second-guessing themselves if they should be relying on Google anymore because this is not working.
Image: miasato.2 / Threads
We do not and cannot blame anyone right now because when a search engine says Batman is a cop then there’s not a lot to expect from it, right?
Clearly, Google had a lot of explanations to provide for the behavior before things really went downhill. The company’s rep claims the errors came when fairly uncommon questions were thrown in the search engine’s direction and therefore it was wrong to state that it represented the majority of users’ experience.
Wow, Google really needs to do better than that if it wishes to continue being the leader in search because stuff like that isn’t going to cut it with users. But Google did admit that it’s working on taking action against policies and any violations carried out against it. Moreover, it hopes to better refine the entire search product by taking such examples or shall we say errors into consideration.
Google never promised that AI would be perfect and that humans cannot blatantly rely on the technology. They even went as far as to call the whole plan experimental when rolling out AI-generated responses too. However, from what we can see, it’s quite clear how such tools are not ready to provide the right data at scale.
The fact that Google made the launch of AI Overviews such a huge deal at this year’s I/O had people excited. Those who attended were given the biggest demo that left a lot of queries in viewers' minds about generative AI being tomorrow’s leader when it’s yet to be at par with human minds.
But remember, it’s not just the Android maker that’s trying its best to grapple with hallucinations linked to the world of AI. Other tech industry names like Perplexity and even Meta are trying their best.
The reason why Google is getting called out so much is because it’s the first from the list to roll out the tech on a scale as grand as it is right now so you’ll find plenty of examples to relate to on this front.
Remember, firms that are involved in the likes of AI creation at a quick pace tend to take major accountability for systems and the approach they are abiding by. It’s a huge responsibility on Google’s shoulders but at the end of the day, even Google knows that it has limited control in terms of what AI might come up with. It’s uncontrollable sometimes!
So what can we say other than AI has taken over Google and slowly the world? We might be on the verge of something super exciting but issues like these keep growing as technology becomes more and more advanced.
Read next: TikTok Adds Restrictions On State-Backed Media To Combat Covert Influence Campaigns
Not only is the offering giving out inaccurate replies at the speed of light but there’s zero accountability from the company with no plans of getting things fixed. Did we mention ads are being included too and it appears that it’s giving rise to huge frustrations across the board?
Turning to the search engine giant to get a great response is just mind-blowing for many to comprehend because Google is doing the exact opposite of what it’s supposed to do.
We agree the feature isn’t toggled to interfere in every single reply but the fact that it goes through the web to produce AI-based replies has people talking. So far, we have seen one query about pizza glue make the rounds and the reply generated was taken from an ancient thread posted across Reddit nearly 10 years back. Mind you, that was a joke on the app but AI Overviews really did take it seriously.
This happens to be one of the long list of mistakes popping up through the latest feature online which Google published in May. But it’s the start of a huge list of wrong answers that has people second-guessing themselves if they should be relying on Google anymore because this is not working.
Image: miasato.2 / Threads
We do not and cannot blame anyone right now because when a search engine says Batman is a cop then there’s not a lot to expect from it, right?
Clearly, Google had a lot of explanations to provide for the behavior before things really went downhill. The company’s rep claims the errors came when fairly uncommon questions were thrown in the search engine’s direction and therefore it was wrong to state that it represented the majority of users’ experience.
I saw someone else mention this one on another post
— Raven - 🍉 (@RavenTenebris13) May 20, 2024
Ah yes , the snake very famous for being a mammal 🤣 pic.twitter.com/xo538FcqoK
Wow, Google really needs to do better than that if it wishes to continue being the leader in search because stuff like that isn’t going to cut it with users. But Google did admit that it’s working on taking action against policies and any violations carried out against it. Moreover, it hopes to better refine the entire search product by taking such examples or shall we say errors into consideration.
Google never promised that AI would be perfect and that humans cannot blatantly rely on the technology. They even went as far as to call the whole plan experimental when rolling out AI-generated responses too. However, from what we can see, it’s quite clear how such tools are not ready to provide the right data at scale.
Seems the origin of the Google AI’s conclusion was an 11 year old Reddit post by the eminent scholar, fucksmith. https://t.co/fG8i5ZlWtl pic.twitter.com/0ijXRqA16y
— Kurt Opsahl @kurt@mstdn.social (@kurtopsahl) May 23, 2024
The fact that Google made the launch of AI Overviews such a huge deal at this year’s I/O had people excited. Those who attended were given the biggest demo that left a lot of queries in viewers' minds about generative AI being tomorrow’s leader when it’s yet to be at par with human minds.
But remember, it’s not just the Android maker that’s trying its best to grapple with hallucinations linked to the world of AI. Other tech industry names like Perplexity and even Meta are trying their best.
The reason why Google is getting called out so much is because it’s the first from the list to roll out the tech on a scale as grand as it is right now so you’ll find plenty of examples to relate to on this front.
Remember, firms that are involved in the likes of AI creation at a quick pace tend to take major accountability for systems and the approach they are abiding by. It’s a huge responsibility on Google’s shoulders but at the end of the day, even Google knows that it has limited control in terms of what AI might come up with. It’s uncontrollable sometimes!
So what can we say other than AI has taken over Google and slowly the world? We might be on the verge of something super exciting but issues like these keep growing as technology becomes more and more advanced.
Read next: TikTok Adds Restrictions On State-Backed Media To Combat Covert Influence Campaigns
