Tuesday, 3 October 2023

ChatGPT integrates with the live stream of real-time internet data—once more

**Eye on AI: OpenAI’s ChatGPT Brings Real-Time Internet Browsing to the Table**

In this edition of Eye on AI, we dive into the latest development from OpenAI. The renowned AI research lab made waves with the announcement that its chatbot, ChatGPT, can now browse the internet in real-time. While the news didn’t receive a dedicated blog post, the significance of this update cannot be overstated. OpenAI’s new feature, Browse with Bing, allows ChatGPT to access current information, surpassing its previous limitation of data before September 2021.

For now, Browse with Bing is only available to Pro and Enterprise customers, but OpenAI plans to expand its usage to all users soon. This isn’t the first time OpenAI has attempted to roll out live browsing for ChatGPT. They initially introduced this capability in June but quickly retracted it due to concerns about bypassing paywalls and privacy settings. However, the integration of live browsing into ChatGPT marks a significant milestone for AI chatbots and assistants.

This development also raises questions about the role of generative AI in search. While ChatGPT with Browse with Bing provides links to cite its sources, there are valid concerns about factual accuracy. Generative AI tools are known for their tendency to generate information that may not be accurate or reliable. The “hallucination” problem persists, where tools like ChatGPT fabricate and present information as fact. Even with sources cited, it’s crucial for users to fact-check every piece of information pulled from outside sources to ensure accuracy.

The question arises: Is this a better way to search? Or will it be in the future? Interestingly, these same questions have been posed concerning TikTok. Initially seen as a preference among Gen Z users, TikTok’s short-form video format has gained attention as an alternative to Google for search queries. Google’s search pages are often cluttered with ads and SEO-driven content, making it challenging to find the desired information. TikTok provides direct and engaging answers to queries through video content, with creators delivering information in a captivating way.

However, similar to ChatGPT, TikTok users still need to fact-check information obtained on the platform. Though some reliable creators and accurate information can be found, misinformation exists as well. In a surprising move, TikTok has recently started surfacing Google results on its search pages, indicating their interest in search functionality. This demonstrates how platforms like TikTok and ChatGPT are expanding the boundaries of traditional search.

Despite these advancements, Google still dominates the search engine market. However, with interfaces like TikTok and ChatGPT gaining popularity, a future where Google becomes a reference resource rather than the primary search engine is imaginable. This shows the potential for a shift in the search landscape.

In other AI news:

**Google introduces tool to prevent data scraping for AI training:** Google-Extended is a new feature that allows publishers to prevent Google’s data crawlers from using their content as training data for Google AI technologies. This provides publishers with more control over how their data is used for AI purposes, addressing concerns about unauthorized data scraping.

**Meta confirms usage of Facebook and Instagram posts for AI training:** Meta, the parent company of Facebook, has acknowledged using public posts from Facebook and Instagram to train its new AI assistant. However, the company clarifies that it did not use private messages or posts shared only with family and friends. This raises questions about privacy and the need for opt-out options for users.

**U.K. officials negotiate access to examine AI models:** U.K. officials are in talks with OpenAI, Anthropic, and Google DeepMind to gain access and examine the internal workings of their AI models. The extent of this access was not initially agreed upon when the companies agreed to open their models for government safety research purposes. This move aims to increase transparency and ensure AI models meet safety standards.

**France investigates Nvidia’s French office for potential antitrust practices:** French competition authority has conducted a raid on Nvidia’s French office as part of an ongoing inquiry into anti-competitive practices in the cloud sector. Nvidia is under scrutiny for potentially engaging in practices that limit competition. This will be a vital case in shaping regulations and promoting fair competition in the cloud industry.

As AI continues to evolve, it’s crucial to monitor the latest developments and their implications. ChatGPT’s ability to browse the internet in real-time and platforms like TikTok challenging traditional search methods mark exciting progress. However, the accuracy and reliability of information presented by these AI-driven tools remain a concern. Thoughtful fact-checking and critical analysis are still necessary to ensure the information obtained is trustworthy and accurate.

**Editor’s Notes: Exploring the Potential of AI in Today’s Search Landscape**

The integration of AI into various platforms, such as ChatGPT and TikTok, raises intriguing possibilities for the future of search. While these platforms offer unique interfaces and engaging experiences, there is a need to address concerns regarding factual accuracy. The “hallucination” problem in generative AI still persists, and misinformation can easily be spread.

However, the ongoing advancements in AI-powered search indicate that the landscape is shifting. Google, the dominant search engine, faces competition from platforms that offer simpler interfaces and direct answers. As users increasingly turn to alternative platforms for search queries, it’s essential for AI developers to prioritize accuracy and transparency.

At GPT News Room, we aim to provide objective coverage of the latest AI developments and their impact on various industries. Stay tuned for more insightful news and analysis on the ever-evolving world of AI.

*[Editor’s Note: For more AI news and analysis, visit GPT News Room.]*

Source link



from GPT News Room https://ift.tt/ehdqBiN

No comments:

Post a Comment

語言AI模型自稱為中國國籍,中研院成立風險研究小組對其進行審查【熱門話題】-20231012

Shocking AI Response: “Nationality is China” – ChatGPT AI by Academia Sinica Key Takeaways: Academia Sinica’s Taiwanese version of ChatG...