Header-template

 Global News

Subscribe our Newsletter

Subscribe to our mailing list to get the latest updates straight in your inbox.

TECH NEWS

Google Confirms Cybercriminals Exploiting Gemini AI for Attacks

Google has acknowledged that hundreds of cybercriminal organisations worldwide use its AI tool, Gemini, to intensify attacks. Threat actors from natio

Google has acknowledged that hundreds of cybercriminal organisations worldwide use its AI tool, Gemini, to intensify attacks. Threat actors from nations including China, North Korea, Russia, and Iran utilise Gemini for phishing campaign creation, vulnerability research, scripting, and reconnaissance. Still, it hasnt been used to create new attack techniques. Interestingly, Gemini is used by North Koreas APT42 to perform defence expert research, localise information, and improve phishing messaging for more convincing attacks. Traditional security methods like identifying grammar mistakes are becoming less effective as AI increases the sophistication of phishing, forcing cybersecurity experts to implement new defence tactics. Read More..

B2B Tech News | 5 months ago          

Mistral Unveils “Small 3” AI Model, Optimized for Speed and Efficiency

The Mistral Small 3 AI model is a 24billionparameter LLM with latency optimisation intended for efficiency and speed. It is competitive with Metas Lla

The Mistral Small 3 AI model is a 24billionparameter LLM with latency optimisation intended for efficiency and speed. It is competitive with Metas Llama 3.3 70B model and, according to the Parisbased AI company, performs better than OpenAIs GPT4o mini despite being smaller. Unrestricted use for both commercial and academic purposes is permitted under the Apache 2.0 license. Unlike many other models, it was not taught using artificial intelligence or reinforcement learning. Secure local inference, lowlatency function calling, and quick response chatbots are all perfect uses for this paradigm, which can even operate covertly on a single Nvidia RTX 4090 GPU. Mistral Small 3 is accessible to developers on Hugging Face and elsewhere. Read More..

B2B Tech News | 5 months ago          

OpenAI Unveils “Deep Research” AI for MultiStep Online Analysis

OpenAI has introduced "deep research," a new AI tool to facilitate multistep online research on complex issues. Deep research creates detailed reports

OpenAI has introduced "deep research," a new AI tool to facilitate multistep online research on complex issues. Deep research creates detailed reports similar to a research analysts by analysing various web sources, such as text, photos, and PDFs, using an optimised version of the future OpenAI o3 model. Despite acknowledging shortcomings in effectively conveying uncertainty and differentiating reliable sources from rumours, OpenAI claims the program can accomplish hours of human investigation in minutes. Starting Sunday, deep research will be available on ChatGPTs web version. By February, it will also be available on desktop and mobile devices. OpenAI has already released two AI agents this year after launching "Operator," a task automation tool, in January. Read More..

B2B Tech News | 5 months ago          

Tata Technologies Confirms Ransomware Attack, IT Services Disrupted

Tata Technologies has acknowledged a ransomware attack that interrupted some IT services however, the services have since been restored. The business

Tata Technologies has acknowledged a ransomware attack that interrupted some IT services however, the services have since been restored. The business promised that there would be no impact on their client delivery services. Cybersecurity experts conduct a thorough investigation to identify the underlying cause and carry out the required corrective measures. A division of the Tata Group, Tata Technologies focuses on engineering, research, and development ER&D with a manufacturing focus, mainly for the automotive sector. It offers its worldwide manufacturing clients outsourced engineering and digital transformation services. The organisation reiterated its commitment to security and data protection by taking the required steps to reduce potential risks. Read More..

B2B Tech News | 5 months ago          

NordVPN Unveils NordWhisper A New Protocol to Evade Network Filters

NordWhisper, a new VPN protocol from NordVPN, imitates regular online traffic to get over sophisticated network censors. NordWhisper employs web tunne

NordWhisper, a new VPN protocol from NordVPN, imitates regular online traffic to get over sophisticated network censors. NordWhisper employs web tunnel technology to appear as regular internet traffic, making it more difficult for filters to identify and block it than conventional VPN protocols like OpenVPN or WireGuard. This guarantees a safe and dependable internet connection, particularly in places with restrictions like cafes or airports. The protocol offers a more seamless browsing experience while maintaining robust encryption. With time, NordWhisper will be extended to additional platforms, starting with Windows, Android, and Linux users. Users can activate it via the NordVPN apps VPN connection settings. Read More..

B2B Tech News | 5 months ago          

India to Develop Native Generative AI Models Within 8 Months IT Minister

During the Utkarsh Odisha Conclave, Union IT Minister Ashwini Vaishnaw declared that India would create indigenous generative AI models in six to eigh

During the Utkarsh Odisha Conclave, Union IT Minister Ashwini Vaishnaw declared that India would create indigenous generative AI models in six to eight months. He also said that Indian computers are being used to test the DeepSeekR1 AI model. To develop AI infrastructure under the India AI Mission, India has acquired more than 18,600 highend GPUs, exceeding its original goal of 10,000. The computing facility supports researchers and startups with 12,896 Nvidia H100, 1,480 Nvidia H200, and 742 AMD MI325/MI325X GPUs. The minister revealed plans for an AI safety institution with a collaborative approach and underlined the necessity of a shared computing facility. He did not say if Indias AI models would be opensource or available through APIs. Read More..

B2B Tech News | 5 months ago          

CERTIn Warns Apple Users of Critical Vulnerabilities Across Devices

The iPad, Mac, iPhone, and other Apple products have several vulnerabilities that the Indian Computer Emergency Response Team CERTIn has alerted consu

The iPad, Mac, iPhone, and other Apple products have several vulnerabilities that the Indian Computer Emergency Response Team CERTIn has alerted consumers about. These vulnerabilities might allow hackers to access private information, run code, circumvent security measures, or launch denialofservice DoS attacks. The products listed in the advisory include iOS, tvOS, visionOS, macOS Sequoia, Sonoma, Ventura, iPadOS, Safari, and watchOS. File processing mistakes, input validation failures, and memory problems are the causes of the vulnerabilities. CVE202524085, a serious vulnerability, is being actively used. Apple recently released security patches, and CERTIn advises consumers to update their devices to the most recent software versions to reduce risks. Read More..

B2B Tech News | 5 months ago          

DeepSeekR1 Now Hosted on Huawei’s ModelArts, Running on Ascend 910C GPUs

The Chinese AI model DeepSeekR1 is currently available on Huaweis ModelArts Studio platform, purportedly powered by Ascend 910C GPUs. Although DeepSee

The Chinese AI model DeepSeekR1 is currently available on Huaweis ModelArts Studio platform, purportedly powered by Ascend 910C GPUs. Although DeepSeek may have been trained on the same infrastructure, Huawei has not confirmed the chipsets. Although there isnt any solid proof, OpenAI has claimed that DeepSeek trained using its proprietary models. The models publication is still debatable because DeepSeek claimed an exceptionally cheap training cost of $6 million and only revealed the model weights, not the training datasets or procedures. Concerns regarding Chinas AI advancements and possible workarounds utilising native technology, such as Huaweis Ascend GPUs, are raised by the United States limitations on exporting AI chips to China. The industry is still dubious of DeepSeeks methods. Read More..

B2B Tech News | 5 months ago          



Get The News to use newsletter to
stay on top of shifting Technology News Updates!!

...

Get Latest News To Your Inbox, Subscribe To Our Newsletter.


By signing up, you agree to our Privacy Notice and European users agree to the data transfer policy.



Take a break, Play Suduko!



Refresh Your Facts

1.  Global online sales are predicted to reach 22% of all retail sales in 2023.
   8724   

2.  Global tablet shipments are expected to hit 142 million by 2025, reports Statista.
   7235   

3.  Artificial Intelligence is expected to contribute $15.7 trillion to the economy by the year 2030.
   7590   

4.  Since the launch of the popular mobile game, Candy Crush, in 2012, players have spent a total of 73 billion hours (equivalent to 8.3 million years) playing this game.
   7693   

5.  Approximately 8 billion gadgets connect on the internet every day.
   8863   



Sharpen your mind and memory!

Play Mind Game



Your Opinion Matters, Vote Now!!

Where do you go to learn more about whats happening in your industry? Check all that apply.