Technology media highlights Technology media highlights

Google launches its anticipated AI search feature, but it’s not all smooth sailing.

In mid-May, Google launched its much-anticipated generative AI search feature to the general public. This new feature gives users an AI-generated overview of their search results at the top of the page, with the goal of creating a more efficient search experience. However, this shift hasn’t gone as smoothly as the tech giant might have hoped. 

Right off the bat, Google users noted that the AI overview feature sometimes provided inaccurate and sometimes nonsensical results. This is partially because the AI is often unable to tell the difference between jokes and facts, meaning that Reddit memes can sometimes make their way into search results. 

Google is already taking action to remove inaccurate search results, particularly those that could be dangerously misleading. However, many users are still concerned about the feature and are looking for ways to remove AI overviews from their personal search results. Businesses are also concerned about how AI overviews will change search engine optimization practices. 

This is just the first step in Google’s rollout of AI-based search features. In the future, the platform plans to add advertisements to AI-generated overviews. Critics worry that this could make it even more difficult for users to find the information they’re looking for. 

This is just one of a myriad of challenges that Google is currently facing. This is just one of a myriad of challenges that Google is currently facing. Over 2,500 internal documents were recently leaked, some of which may have contained information about the company’s proprietary search algorithms.

ChatGPT has already had a turbulent year, and the generative-AI company finds itself under the microscope yet again with the launch of its new conversational AI tool. This tool uses five voices to provide generative AI responses via audio. 

However, users quickly noticed that one of the voices, named Sky, was eerily reminiscent of Scarlett Johansson’s character in the 2013 film Her. In the movie, Johansson plays an AI voice assistant-turned-girlfriend. 

Johansson promptly hired legal counsel, prompting ChatGPT to take down the “Sky” voice. According to a statement released by Johansson, ChatGPT CEO Sam Altman had previously asked her to voice the forthcoming AI assistant, but the actress declined. 

Records show that Sky is not a voice clone of Johanssen, but was instead voiced by a different actress, who hasn’t been named due to privacy concerns. However, the situation has raised broader concerns about AI-generated likenesses and privacy among technology enthusiasts. This could potentially lead to changes in IP legislation in the future. 

This debacle comes after a slew of OpenAI employees have left the company amid concerns about the product’s long-term safety. In response, OpenAI has assembled a new safety committee. However, this committee consists entirely of internal leaders and board members, without any external oversight.

2024 is a significant year for global elections, as millions of people around the world will be voting to choose new government leaders. This includes notable elections in the United States, India, Mexico, Russia, and the European Union. 

Unfortunately, the risk of election interference is high. Recent advances in technology have made it easier than ever for malicious entities to influence public sentiment and potentially sway the outcome of major elections. So far this year, we’ve already seen many instances of potential election interference as a result of AI-generated deepfakes and social media campaigns. 

One of the most notable examples was an AI-generated robocall during the New Hampshire primary that impersonated President Joe Biden. The call was sent out to Democratic voters, urging them not to participate in the primary. This has prompted the FCC to take action, outlawing the use of AI-generated voices in robocalls. 

In India, politicians are taking a very different approach to AI technology. Many Indian politicians are creating AI video and audio deepfakes of themselves ahead of the upcoming elections, using the technology to “talk” with thousands of voters across the country. However, this has made it difficult for voters to tell which deepfakes are sanctioned by politicians and which have been used for political interference. 

Russia has already had its own presidential elections this year, and now groups aligned with the Kremlin are using online campaigns to spread misinformation about politics in the United States and other western countries. Kremlin-backed interference also happened in the lead-up to the 2016 and 2020 presidential elections in the US. 

Current Russian misinformation campaigns are focused primarily on college campus protests. These campaigns create lookalike versions of popular online news outlets with fake articles that contain misinformation. These articles are then spread via social media using bots.

It’s been nearly 20 years since the birth of social media, and so much has changed in this time. The social media landscape feels particularly volatile right now, as Elon Musk has transformed Twitter into X and the US government approved a potential TikTok ban. 

With so many ups and downs, social media platforms have to consistently add new features and functionality to keep their audiences engaged. In May alone, we’ve seen many top social media networks announce important changes and innovations. 

For example, YouTube has recently added a collection of free games to its platform, which will soon become available to all users. These games could potentially serve as a new stream of ad revenue and attract new users. Snapchat has also added a plethora of new features lately, including the ability to edit messages and send Bitmoji reactions to friends. Snapchat’s also adding an AI chatbot to its messaging features. 

Tumblr is also making a pivot after more than a decade of ups and downs. The platform announced that it will launch a new Communities feature, which will function similarly to Reddit. This could potentially help Tumblr compete with larger social media platforms. 

Amazon has been developing and testing drones as an option for delivery since 2013. In 2022, Amazon launched Prime Air test programs in Lockeford, California and College Station, Texas. The Lockeford program shut down in April 2024, with a new location opening in Tolleson, Arizona. The company also plans to expand the Prime Air testing program to Italy and the United Kingdom. 

These tests use drones to deliver products over very short distances, but previously didn’t allow drones to fly outside of the controller’s line of sight. However, a new FAA approval will allow the drones to fly outside of the line of sight thanks to onboard detect-and-avoid systems. Amazon quickly announced that they would be expanding their delivery program in College Station as a result, with a broader goal of making 500 million drone deliveries by 2030.

Drones would help Amazon reduce its reliance on delivery drivers and make it easier to keep up during periods of high demand. Drones could also be used to deliver products to remote areas without efficient road access.