Cybercriminals Shift Focus to Credential Theft and AI-Driven Phishing

Article Highlights
Off On

In 2024, the cybersecurity landscape experienced notable changes, with cybercriminals increasingly adopting stealthier tactics, particularly focusing on credential theft through infostealers. IBM’s research highlighted an 84% spike in phishing emails delivering infostealers from 2023 to 2024, favoring this method for its quick and discreet identity attacks.

Credential theft incidents surged as attackers exploited the growing complexity of hybrid cloud environments, targeting critical infrastructure. Within this sector, cybercriminals often gained initial access via valid accounts (31%), phishing (26%), and public-facing application exploits (26%). Once access was secured, malware was deployed in 40% of cases, with ransomware constituting 30% of these deployments.

A significant trend in 2024 saw cybercriminals prefer stealing data (18%) over encrypting it (11%). This shift is attributed to advanced detection technologies and increased pressure from law enforcement, prompting quicker and more discreet operations. Although ransomware incidents declined, constituting 28% of malware cases and 11% of security cases, cybercriminals continued diversifying their malware portfolios.

IBM’s report exposed vulnerabilities in critical infrastructure due to reliance on legacy technology and slow patching cycles. Over a quarter of incidents exploited these weaknesses. The trading of exploit codes for common vulnerabilities on dark web forums fueled attacks on power grids, health networks, and industrial systems. Phishing emails delivering infostealers continued to rise into early 2025, with a 180% increase compared to 2023. The use of AI enabled cybercriminals to scale phishing campaigns, making identity attacks cheaper and more profitable. Additionally, the dark web saw an increase in adversary-in-the-middle (AITM) phishing kits designed to bypass multi-factor authentication (MFA).

Regionally, the Asia-Pacific (APAC) region accounted for 34% of attacks, focusing primarily on malware-ransomware (22%) and server access (11%). North America followed with 24% of incidents, with common actions including remote access tools (17%) and backdoor malware (17%). AI security remained a critical concern as researchers worked to identify and rectify vulnerabilities before exploitation by cybercriminals. As AI adoption grows, securing the AI pipeline will be crucial. In conclusion, 2024 marked significant shifts towards stealthier and more opportunistic cyberattack methods. The emphasis on credential theft and data exfiltration shows cybercriminals adapting to modern security measures. This evolution underscores the importance of robust and proactive cybersecurity measures to combat an ever-evolving threat landscape.

Explore more

How Career Longevity Can Stifle Your Professional Growth

The traditional belief that a long and stable tenure at a single organization serves as the ultimate hallmark of a successful career has begun to crumble under the weight of rapid industrial evolution. While many professionals historically viewed a decade in the same office as a badge of honor, the modern landscape suggests that this perceived stability might actually be

The Hidden Risks of Treating AI Like a Human Colleague

Corporate boardrooms across the globe are currently witnessing a fundamental transformation in how digital intelligence is integrated into the traditional workforce hierarchy. Rather than remaining relegated to the background as specialized software, artificial intelligence is now being personified as a dedicated teammate with a specific identity. Recent industry data indicates that approximately 31% of leadership teams have started framing AI

Why People and Data Are the Real Keys to NetDevOps Success

While the modern enterprise landscape is saturated with powerful Python libraries and sophisticated Ansible playbooks, the actual transformation of network infrastructure often remains trapped within the confines of isolated lab environments. The promise of “push-button” infrastructure has existed for years, yet many organizations find their NetDevOps initiatives stalled. This stagnation is rarely the result of a missing software capability or

When Should DevOps Agents Act Without Human Approval?

The catastrophic failure of a global banking system caused by a single misconfigured automation script remains the industry’s ultimate cautionary tale, haunting every engineer who contemplates pressing the ‘enable’ button on a fully autonomous AI agent. While the promise of self-healing infrastructure has existed for years, the transition from human-managed workflows to agent-led systems is fraught with psychological and technical

GitHub Spec Kit Replaces Vibe Coding with Precise Engineering

The days of tossing vague sentences into a chat box and hoping for functional code are rapidly coming to an end as software engineering demands a move toward verifiable precision. This shift is becoming necessary because the novelty of generative AI is wearing off, revealing a landscape littered with “hallucinated” logic and architectural inconsistencies. The arrival of GitHub’s Spec Kit