Google has announced that its Deep Research feature, part of its Gemini AI program, is now available in over 45 languages, including Arabic, in 150 countries. This feature, launched earlier this month, allows users of the premium Google One AI plan to take advantage of an AI-powered search tool.
The supported languages for "Gemini" include Arabic, Bengali, Chinese, Danish, French, German, Gujarati, Hindi, Indonesian, Italian, Japanese, Kannada, Korean, Malayalam, Marathi, Polish, Portuguese, Swahili, Spanish, Tamil, Telugu, Ukrainian, and Urdu.
With Deep Research, Gemini Advanced users can create in-depth research reports filled with sources and references. According to Google, "With Deep Research, you can create comprehensive research reports on complex topics, with links to sources, saving you hours of research in just minutes." Users can also request further clarifications on specific points or modify the report, with the ability to export the document to Google Docs in a text file.
However, the challenge for the company lies in finding reliable sources in a specific language and then summarizing them in the native language without causing linguistic errors.
Controversial Changes in Gemini's Guidelines
On the other hand, new guidelines issued by Google for its contractors, who evaluate AI responses in "Gemini," have raised concerns about the accuracy of the information provided by the program.
Contractors are now prohibited from skipping requests that require specialized knowledge, even if the topic is outside their area of expertise. This change could lead to inaccurate evaluations of sensitive topics, such as healthcare.
The program has seen a shift in its guidelines. Previously, contractors were allowed to skip requests they didn't have sufficient knowledge to assess, according to a report by "TechCrunch." However, the new guidelines now require contractors to evaluate requests regardless of their expertise, with a note indicating the lack of knowledge.
Experts fear that this policy could lead to the dissemination of inaccurate information, particularly on sensitive topics such as rare diseases and healthcare. Many contractors have expressed concerns about this change, noting that the original policy was intended to ensure accurate evaluations by qualified individuals.
Gemini's Launch and Advancements
In December 2024, Google launched its Gemini AI system, which competes directly with OpenAI's GPT-4 model, focusing heavily on multi-modal capabilities and diverse functionality. According to Google, Gemini was designed to be multi-modal from the start, being trained on various types of data. This diversity enables Gemini to understand and process different inputs like text, code, audio, images, and video seamlessly.
The launch of Gemini marks an important step for Google in the AI race, highlighting innovation and responsibility. This advancement showcases Google's ability to challenge existing systems and offer advanced technological solutions that meet users' diverse needs.
Stay tuned for more updates on Google's efforts to fight spam and protect against breaches.
Post a Comment