Breaking
RUAn explosion occurred in the city of Izyum, Kharkov region.ITRussian attacks on Kiev and other Ukrainian cities: casualties and infrastructure damageRUUlyanovsk was subjected to a massive UAV attack: there were casualtiesINTLRussian drone attack on Kyiv kills 14-year-old boy as Denmark warns of Nato risksRUNational Park "Krasnoyarsk Pillars" introduces a complete ban on visits due to bearsTRPress Statement from the Families in the Ship Disaster Case in KyreniaRUA gang of women who profited from SVO participants was detained in YekaterinburgRUNetanyahu's plane flew into Canadian airspace twiceCRYPTO-ENSamourai Wallet Co-Founder Keonne Rodriguez Faces Prison TransferPLThe longest bridge in Podkarpacie is taking shapeRUAn explosion occurred in the city of Izyum, Kharkov region.ITRussian attacks on Kiev and other Ukrainian cities: casualties and infrastructure damageRUUlyanovsk was subjected to a massive UAV attack: there were casualtiesINTLRussian drone attack on Kyiv kills 14-year-old boy as Denmark warns of Nato risksRUNational Park "Krasnoyarsk Pillars" introduces a complete ban on visits due to bearsTRPress Statement from the Families in the Ship Disaster Case in KyreniaRUA gang of women who profited from SVO participants was detained in YekaterinburgRUNetanyahu's plane flew into Canadian airspace twiceCRYPTO-ENSamourai Wallet Co-Founder Keonne Rodriguez Faces Prison TransferPLThe longest bridge in Podkarpacie is taking shape
Newsgather
All StoriesWorldSportsFinanceTechScienceHealthCultureClimatePoliticsSpace
BackGoogle Completed the Development Phase of Gemini 4 Artificial Intelligence Model
Google Completed the Development Phase of Gemini 4 Artificial Intelligence Model
Developing
ShiftDelete1 hour agoTech2 min readTürkiyeView original

Google Completed the Development Phase of Gemini 4 Artificial Intelligence Model

googleGemini 4artificial intelligence

Quick Look

  • Google DeepMind manager Koray Kavukcuoglu announced that the development process of the Gemini 4 artificial intelligence model has ended and security tests are continuing.
  • It was stated that the model will be released before the end of 2026.

AI-generated summary

Why It Matters

Google is being compared to companies like OpenAI and Anthropic in the artificial intelligence competition. Gemini series are Google's strategic products in this field.

Font size

Google is nearing the end of the development phase of its new flagship artificial intelligence model, Gemini 4. Here are all the details about the release date of the model and the company's new strategy.

Google DeepMind manager Koray Kavukcuoglu announced that the development process of the company's next flagship artificial intelligence model, Gemini 4, is nearing completion and that the model will be released as soon as possible. Speaking at an event where the artificial intelligence agenda was discussed, Kavukcuoglu stated that the model is currently in the post-training process and security tests are continuing.

In response to the rapid innovation processes of its competitors, Google aims to maintain its leadership position in the world of artificial intelligence with Gemini 4. The company stated that it is very hopeful about the performance results of the model, and although it refrained from specifying an exact release date, it emphasized that a launch is expected much before the end of 2026.

Development Process Continues Unabated

Artificial intelligence studies on Google's side are of strategic importance, especially in this period when competition is intensifying. Koray Kavukcuoglu stated that Gemini 4 is currently being actively used in testing processes and intense efforts are being made to make the behavior of the model compatible with the end user experience.

Creating firewalls and checking the model against possible risks are the most critical parts of this stage.

Aiming for Leadership in a Competitive Environment

New updates offered by competitors such as OpenAI and Anthropic in recent months have caused speculation in the industry that Google is lagging behind. However, Kavukcuoglu expressed his belief that Google will always remain a pioneer in the artificial intelligence race and stated that he has full confidence in its teams. The company made a strategic choice by directing its resources to the more powerful Gemini 4 instead of working on Gemini 3.5 Pro.

Artificial Intelligence Focus Shifts

Google's vision is not just to develop more powerful models, but also to build reliable and intelligent artificial intelligence agents. Kavukcuoglu argues that it is more important to go beyond discussions of artificial general intelligence (AGI) and create autonomous systems that people can trust. Gemini 4 is Google's most comprehensive response to this goal.

What to Watch

AI outlook — possibilities, not facts

  • The Gemini 4 model will be released at the end of 2026.

    Likely · Within months

Open Questions

  • When will the exact release date of Gemini 4 be?
  • How good will the model's performance benchmarks be?
  • Which sectors does Google target with Gemini 4?

Related Topics

google
PeopleKoray Kavukcuoglu
OrganizationsGoogleGoogle DeepMindOpenAIAnthropic
PlacesUnited States
TopicsGemini 4artificial intelligencekoray kavukcuogludeepmindsecurity testscompetition
This article was originally published by ShiftDelete.

Related Stories

EMRA increased its power to install integrated storage facilities at charging stations
Developing·

EMRA increased its power to install integrated storage facilities at charging stations

The Energy Market Regulatory Authority (EPDK) has increased the power at which integrated electricity storage facilities can be installed at charging stations to a level higher than the agreed power of the charging station. This arrangement makes it possible to benefit from direct storage systems in high-power charging stations without making a network investment. Automatic power control and cut-off systems were made mandatory for network security.

AA Güncel
2 min read
Meta's Muse AI Assistant Makes Music and Podcast Management Easier with Spotify Integration
Developing·

Meta's Muse AI Assistant Makes Music and Podcast Management Easier with Spotify Integration

Meta's artificial intelligence assistant Muse makes music and podcast management possible via the chat window, thanks to Spotify integration. Users can now perform operations such as playing music, listening to audiobooks and managing podcasts while chatting with Muse. There is also a 'Personal Podcast' feature that converts your personal notes into audio podcast format. The service is currently available in the US, Canada and Mexico.

ShiftDelete
2 min read
Google is Preparing to Launch a Prototype Satellite to Test Artificial Intelligence Chips in Space
Developing·

Google is Preparing to Launch a Prototype Satellite to Test Artificial Intelligence Chips in Space

As part of Project Suncatcher, Google will test the durability of TPU chips in space conditions by sending a prototype satellite to low Earth orbit with SpaceX's Transporter-18 launch. The ability of AI chips to withstand launch forces, radiation and temperature changes aims to evaluate the technical limits of the idea of ​​​​establishing a data center in space.

NTV Teknoloji
2 min read
Meta's New Ray-Ban Gen 3 and Smart Glasses with/Without Camera Tested
Developing·

Meta's New Ray-Ban Gen 3 and Smart Glasses with/Without Camera Tested

Meta introduced the third generation Ray-Ban Meta Gen 3 smart glasses, developed in collaboration with Ray-Ban, and the Meta Glasses Sound model without a camera. The Gen 3 model offers a multi-modal artificial intelligence experience such as object recognition and instant translation in Turkish via Meta AI, with its improved Qualcomm chipset, reinforced microphone set and long battery life. Glasses Sound, which does not have a camera, weighs 43 grams and provides a privacy-oriented voice assistant and music experience with 12 hours of battery life. While the accessibility potential for the visually impaired is emphasized, it is stated that sales in Türkiye have not started yet and Turkish support is limited in some AI services.

ShiftDelete
2 min read
More on this topicgoogle