Yazılar

X Reportedly Testing Free Grok AI Version With Updated Logo

X (formerly known as Twitter) is reportedly experimenting with a free version of its artificial intelligence chatbot, Grok. Over the weekend, several users claimed to have gained access to Grok without subscribing to the platform’s paid X Premium tier. However, as per user feedback, the free version has certain limitations, and its availability appears to be restricted to specific regions. This trial comes shortly after xAI, the Elon Musk-led AI initiative, launched the Grok API and announced incentives for developers to integrate it.

Reports suggest that this move marks an effort to expand Grok’s reach and potentially attract a wider audience. Initially launched in November 2023, Grok was previously available only to paid subscribers. According to posts by app researchers and tech enthusiasts on X, the free tier could allow a broader range of users to experience the chatbot’s capabilities, albeit with reduced functionality compared to the premium version.

A report by TechCrunch indicated that the free Grok trial is currently being tested in New Zealand, with the possibility of extending to other regions. However, availability remains uneven; for instance, users in India and other countries have noted the absence of the feature. At this stage, the specifics of what the free version offers, and how it differs from the premium experience, remain unclear. Additionally, Gadgets 360 could not independently confirm the free version’s rollout.

In addition to testing a free tier, X is also updating Grok’s branding. The chatbot’s original wordmark logo is reportedly being replaced with a socket-like emblem, signaling a fresh visual identity for the AI tool. Whether these changes are part of a larger strategy to boost Grok’s adoption or improve its appeal to developers and end users remains to be seen.

Misinformation on X Amplifies During U.S. Election Cycle, Experts Say

In the lead-up to the 2024 U.S. presidential election, misinformation on X (formerly Twitter), largely stemming from posts by the platform’s owner Elon Musk, has garnered over two billion views this year, according to the Center for Countering Digital Hate (CCDH). Musk’s substantial following of 203 million has reportedly enabled widespread amplification of these claims, intensifying the platform’s influence over election narratives, particularly in battleground states crucial to the outcome.

A report from CCDH highlights that at least 87 of Musk’s posts related to the election were marked as misleading or false by fact-checkers. Experts argue that this massive dissemination of misinformation on X could shape voter perception and increase polarization as Election Day approaches. “X acts as a conduit, spreading content from one social media platform to others like Reddit and Telegram,” explained Carnegie Mellon University professor Kathleen Carley, an expert in disinformation studies.

Despite these concerns, a spokesperson from X defended the platform’s approach, pointing to the Community Notes feature that allows users to provide added context to potentially misleading posts. The spokesperson argued that Community Notes offer a more effective solution than traditional warning labels, as they encourage users to critically assess content.

Since Musk’s acquisition of X, the platform has reduced its content moderation practices, including layoffs that impacted moderation teams. Musk has also publicly backed former President Donald Trump in a close race against Democratic candidate Kamala Harris. This environment of lax moderation, paired with Musk’s extensive influence on the platform, has fostered what experts call “network effects” that help misinformation spread between platforms.

False narratives about election processes in swing states, such as Pennsylvania, have also gained traction on X. Philip Hensley-Robin of Common Cause, a nonpartisan group advocating for fair voting practices, shared concerns over misleading claims in Pennsylvania. He noted that certain users had inaccurately presented legitimate administrative actions, such as flagging incomplete voter registrations, as election interference. “Some posts implied voter fraud, despite election administrators following all procedures to ensure that only eligible votes were counted,” Hensley-Robin stated.

One prominent instance of election misinformation on X involved a fake video suggesting that mail-in ballots for Trump in Pennsylvania were being destroyed. Cyabra, a digital intelligence company focused on disinformation, reported that an X account with 117,000 followers played a critical role in circulating the video. While the platform’s spokesperson noted that X took action against accounts spreading this misleading video, election experts argue that such interventions often occur too late to curb the narrative’s spread.

The widespread nature of election misinformation on X underscores the complex challenges social media poses to U.S. election integrity, leaving platforms and watchdogs struggling to keep up with the rapid proliferation of misleading information.