🚀 Medial Secures Investment on Shark Tank India - Fueling the Future of Professional Social Networking. 🔥

Startup Showcase

Premium Content

Try our Valuation Calculator →

News on Medial

AI companies face growing concerns over cyber threats to large language models

Economic Times

Economic Times · 10m ago

AI companies face growing concerns over cyber threats to large language models

Protecting large language models (LLMs) from cyber threats is a major concern for companies like OpenAI, Microsoft, and Meta. Anand Eswaran, CEO of Veeam Software, highlighted the importance of securing LLMs to prevent malicious actors from polluting them and creating harmful outcomes. Publicly accessible AI models may face corruption issues that lead to discrimination and other negative consequences. Eswaran underlined the significance of safeguarding LLMs to ensure their responsible use in areas like loan approvals. Veeam's survey revealed that a significant number of organizations experienced ransomware incidents and struggled to recover their data even after paying the ransom.

1

Related News

Microsoft says US rivals are beginning to use generative AI in offensive cyber operations

Economic Times

Economic Times · 1y ago

Microsoft says US rivals are beginning to use generative AI in offensive cyber operations

Microsoft has stated that adversaries like Iran, North Korea, Russia, and China are utilizing generative artificial intelligence (AI) for offensive cyber operations. Microsoft, in collaboration with OpenAI, has detected and disrupted threats that exploited AI technology developed by these adversaries. The company believes that exposing these techniques is crucial as rival nations use large-language models to breach networks and conduct influence operations. This development raises concerns about the potential impact on cybersecurity, including enhanced deepfakes and voice cloning.

China is building its own AI at a rapid pace

Business Insider

Business Insider · 1y ago

China is building its own AI at a rapid pace

Chinese tech giants are rapidly developing their own AI models and trying to catch up with their US counterparts like OpenAI. The Chinese government has approved over 40 AI models in the last six months, including 14 Large Language Models (LLMs) in just the past week. Baidu, often referred to as "China's Google," is leading the charge with its AI chatbot Ernie Bot, which has over 100 million users and is considered a competitor to OpenAI's GPT-4 model. However, Chinese companies face both technological and political challenges, as AI models must adhere to strict government regulations and avoid sensitive topics. Additionally, geopolitical tensions, such as restrictions on semiconductor sales, may hinder China's progress in becoming an AI superpower.

Gupshup rolls out multimodal AI agents to fuel AI adoption for enterprises

YourStory

YourStory · 6m ago

Gupshup rolls out multimodal AI agents to fuel AI adoption for enterprises

Conversational AI platform Gupshup has launched a comprehensive AI Agent Library, featuring industry-specific, pre-built agents designed to enhance enterprise operations. Built on large language models, these agents streamline customer experience and operational efficiency across multiple channels like WhatsApp and SMS. The introduction aligns with a growing trend in agentic AI, as companies aim to modernize interactions through advanced language capabilities. Enterprises benefit from personalized, efficient, and language-inclusive AI-driven interactions. Gupshup plans further AI expansions.

Open-source companies are sharing AI free. Can they crack OpenAI’s dominance?

Livemint

Livemint · 1y ago

Open-source companies are sharing AI free. Can they crack OpenAI’s dominance?

As tech giants like OpenAI dominate the artificial intelligence (AI) market, some companies are trying to compete by offering their AI models for free. Open-source AI models appeal to businesses as they provide access to large language models without the need to pay or share data with proprietary vendors. These companies believe they can generate profit by selling business-grade services and applications on top of their open models and by offering additional features and support that enterprises are willing to pay for. However, open-source AI models still face challenges in terms of initial training costs, licensing, and attracting developers.

Operant AI unveils MCP Gateway to secure emerging runtime AI workflows

IndianStartupNews

IndianStartupNews · 2m ago

Operant AI unveils MCP Gateway to secure emerging runtime AI workflows

Operant AI has launched MCP Gateway, a cybersecurity solution designed to protect AI workflows utilizing the Model Context Protocol (MCP). The gateway offers real-time, active-blocking security for AI systems amidst growing adoption of MCP, which enables large language models to securely interact with various technology layers. With features like MCP Discovery, Detections, and Defense, the solution addresses emerging threats such as prompt injection attacks and inter-agent spoofing, particularly as organizations increasingly integrate open-source MCP servers and tools.

Large language models pose growing security risks

Livemint

Livemint · 6m ago

Large language models pose growing security risks

As large language models (LLMs) become more widespread, cybersecurity risks increase. These generative AI models can expose sensitive data and bring unsafe code into businesses. Threats range from inadvertent data exposure to prompt injections by malicious entities. The lack of structure in LLM interactions complicates security, and companies must be diligent about data handling and human oversight. With governmental policy lagging, businesses are encouraged to foster industry-wide cooperation and establish robust security practices.

India Inc shoring up its cybersecurity spends amid spillover threats

Economic Times

Economic Times · 3m ago

India Inc shoring up its cybersecurity spends amid spillover threats

In response to recent India-Pakistan hostilities, large and mid-sized Indian companies are enhancing cybersecurity measures due to a rise in cyber threats. Previously slow to act, they recognize the severe financial impact of cyberattacks. Following advisories and increased threats, businesses, especially in BFSI and tech sectors, are reassessing cyber defenses, leading to a surge in cybersecurity expenditure. This swift action reflects heightened focus on mitigating geopolitical cyber risks, with urgency in finalizing security contracts.

From sci-fi to state law: California’s plan to prevent AI catastrophe

Arstechnica

Arstechnica · 1y ago

From sci-fi to state law: California’s plan to prevent AI catastrophe

California's proposed bill, SB-1047, aims to regulate the safety of large artificial intelligence (AI) models. The bill requires companies behind AI models with significant training costs to implement testing procedures and systems to prevent and respond to safety incidents. These incidents are defined as AI systems enabling critical harms leading to mass casualties or significant damage. While some argue that the bill is necessary to prevent future AI threats, critics suggest that it focuses too much on fictional concerns and could hinder technological advancements.

Factbox-China’s AI firms take spotlight with deals, low-cost models

Livemint

Livemint · 6m ago

Factbox-China’s AI firms take spotlight with deals, low-cost models

China's AI industry is gaining attention with advancements in low-cost and efficient models. Prominent players include ByteDance's Duobao, Alibaba's Qwen, Baidu's Ernie, Tencent's Hunyuan, Zhipu AI's GLM, and Moonshot's Kimi. These companies focus on creating cost-effective large language models (LLMs) and open-source frameworks like the "mixture of experts." Alibaba and Tencent, known for supporting open AI development, have formed strategic partnerships to advance AI capabilities, highlighting China's growing influence in the global AI landscape.

The NSA Warns That US Adversaries Free to Mine Private Data May Have an AI Edge

Wired · 1y ago

The NSA Warns That US Adversaries Free to Mine Private Data May Have an AI Edge

Gilbert Herrera, research director of the US National Security Agency (NSA), discusses the agency's perspective on artificial intelligence (AI) in a recent interview. Herrera acknowledges the surprise and potential of large language models like ChatGPT but explains that the NSA couldn't have created such models due to privacy laws and budget constraints. He highlights the potential uses of commercially available language models in reverse engineering cyber defenses and analysts' support. However, he also mentions the challenges posed by data retention and legal issues. Herrera emphasizes the importance of AI security and the need for protecting models and data from theft and manipulation.

Trackers

Active Indian VC’s

OG Capital Email

With a hands-on approach, OG Capital aims to invest in over 20 promising...

Accel Partners Email

Early and growth-stage investments in disruptive technology companies with...

Early-stage venture capital firm investing in technology startups in India. Focus on...

Access All Trackers

Startup Showcase Winners

June 2025

Helping your parents when you are miles away

The Pit Stop Your Cravings Deserve

The next generation E-commerce platform

Enter Ongoing Startup Showcase

Top Users

Trending News on Medial

Just Months After Promotion, Open ...

Microsoft locks down a building a ...

Gaming Ban Stumps Cricket: Virat ...

Download the medial app to read full posts, comements and news.