Indian cybersecurity firms including Indusface and Astra Security are adopting AI agents built on large language models to accelerate software vulnerability testing from days or weeks to hours, according to The Economic Times. The shift reflects growing attacker speed and AI tools’ emerging ability to identify exploits autonomously, prompting companies to adopt automated testing to keep pace with evolving threats.
Testing Time Acceleration
Large client security assessments that once required four to five days—or up to 20 days for larger applications—now complete within hours, according to Ashish Tandon, CEO of Indusface. This acceleration allows security teams to identify and address vulnerabilities more rapidly as the threat landscape evolves.
Attacker Speed and Vulnerability Growth
The urgency for faster testing is underscored by emerging data on attacker capabilities. CrowdStrike reported that the average attacker breakout time fell to 48 minutes in 2025. Meanwhile, Gartner projects that annual documented vulnerabilities will exceed 1 million by 2030, compared to approximately 277,000 in 2025—a near-fourfold increase.
AI Capability Expansion
Proofpoint, which expanded operations in India last year, noted that AI agents help review thousands of threat alerts daily. This automation addresses a critical challenge: companies face tightening data regulations and a shortage of qualified security analysts. The capability extends beyond alert triage; according to Anthropic, Claude Mythos Preview identified a bug in OpenBSD, an open-source operating system, that had remained undetected for 27 years. The same model achieved a 72.4% success rate in converting known vulnerabilities into working exploits, compared to 14.4% for Opus 4.6, an earlier Anthropic model.
Remediation and Organizational Disparity
While AI accelerates vulnerability discovery, remediation—the process of fixing security issues—remains a bottleneck requiring human review and approval. According to Arctic Wolf, a cybersecurity company, 76% of compromises in its incident response cases involved one or more of 10 known vulnerabilities for which patches were available before exploitation. This gap may widen across the security landscape: larger enterprises equipped with AI-driven detection and remediation capabilities may outpace smaller organizations lacking sufficient staff or budget to manage the volume of identified vulnerabilities.
Disclaimer: The information on this page may come from third parties and does not represent the views or opinions of Gate. The content displayed on this page is for reference only and does not constitute any financial, investment, or legal advice. Gate does not guarantee the accuracy or completeness of the information and shall not be liable for any losses arising from the use of this information. Virtual asset investments carry high risks and are subject to significant price volatility. You may lose all of your invested principal. Please fully understand the relevant risks and make prudent decisions based on your own financial situation and risk tolerance. For details, please refer to
Disclaimer.
Related Articles
Meta is developing an AI assistant named Hatch to rival OpenClaw, completing internal testing by the end of June
According to a May 5 report by the Financial Times, Meta is developing an AI assistant (Hatch) for mainstream consumers, inspired by OpenClaw from OpenAI, with the goal of completing internal testing by the end of June; at the same time, Meta plans to integrate an independent agentic shopping tool into its Instagram service before the fourth quarter of this year.
MarketWhisper49m ago
Cloudflare: Non-Human Traffic Now Majority, x402 Foundation Addresses Web Economics
Cloudflare's Chief Strategy Officer stated that more than half of internet traffic is now non-human, highlighting a shift in web usage patterns driven by AI agents. The company points to the x402 Foundation as a key initiative building infrastructure to support a sustainable digital content economy
CryptoFrontier2h ago
HINT（Hive Intelligence）24小时大涨60.25%
Gate News: On May 6, according to Gate market data, as of the time of publication, HINT (Hive Intelligence) is trading at $0.001695. It has risen 60.25% over the past 24 hours, with a high of $0.0019 and a low of $0.0010577. The 24-hour trading volume is $100.1k. Its current market cap is about $781.4k.
Hive Intelligence is the foundational infrastructure layer for AI agents, providing a unified application programming interface for real-time blockchain data. It removes data fragmentation, enabling AI agents to effortlessly query and interact on-chain. As an institutional-grade crypto market infrastructure, Hive Intelligence provides MCP endpoints, REST A
GateNews2h ago
Anthropic launches 10 financial AI agents, integrating with Microsoft 365 to easily handle finance tasks
Anthropic launches 10 financial AI agent templates, integrating data partners such as Microsoft 365, Moody’s, and D&B, and they can be used as plugins or with automated scheduling. The templates are divided into two categories: research/customer service and financial operations, covering areas such as investment research, financial models, KYC, and more, and they embed Moody’s ratings. Claude Opus 4.7 ranks first in the financial agent benchmark, indicating that talent and regulatory dynamics in the financial industry will be affected.
ChainNewsAbmedia2h ago
Nvidia and ServiceNow Expand Partnership, Introduce Project Arc Desktop Agent
According to Nvidia, the companies expanded their partnership this week to introduce ServiceNow Project Arc, an autonomous desktop agent secured by Nvidia OpenShell and governed by ServiceNow AI Control Tower. The products are aimed at helping enterprises manage and audit AI agents deployed across d
GateNews3h ago
Solana and Google Cloud launch Pay.sh, AI agents can pay with stablecoins
According to Decrypt’s report on May 6, the Solana Foundation partnered with Google Cloud to launch the Pay.sh service, allowing AI agents to pay for on-demand API access costs using stablecoins on Solana, without traditional accounts or subscriptions; for each API call, agents only need to pay a fraction of a cent, with no minimum spend requirement.
MarketWhisper3h ago
