Insights/ Threat Landscape
Why phishing still presents a significant risk
Phishing is still one of the most effective attack vectors. How it has evolved, what the 2026 data shows, and why simulation training still moves the needle.

Despite decades of cybersecurity investment, phishing remains one of the most effective and persistent vectors of cyberattack. As attackers evolve their tactics, increasingly leveraging AI, personalised social engineering, multi-channel delivery, and deepfake techniques, organisations continue to face meaningful risk across every industry. This article looks at why phishing endures, how it has developed over time, what the current data shows, and the role well-designed simulation programs play in building real resilience.
Data sourcesHoxhunt Phishing Trends Report (2026) · KnowBe4 Phishing Threat Trends (2025) · Verizon Data Breach Investigations Report (2025) · IBM Cost of a Data Breach Report (2025) · ASD's ACSC Annual Cyber Threat Report 2024 to 25Figures reflect publicly reported research as of October 2026. Findings evolve. Verify current studies before citing externally.
01The oldest trick still works
What phishing is, and why it still works
At its core, phishing is a form of social engineering in which attackers deceive people into disclosing sensitive information, credentials, or performing an unsafe action. It remains effective because it exploits human psychology, trust, urgency, authority, and convenience, rather than a technical vulnerability.
KnowBe4 found that 82.6% of the phishing emails it analysed between September 2024 and February 2025 showed some use of AI.
Hoxhunt's 2026 Phishing Trends Report saw AI-generated attacks jump roughly fourteenfold late in 2025, from under 5% of reported attacks for most of the year to a peak of 56% in December. AI removes the classic warning signs: clumsy grammar and obvious errors are gone, replaced by fluent, personalised messages produced faster than any human team could write them.
02From spam to spear
The evolution of phishing: from generic emails to multi-channel attacks
Phishing has evolved far beyond the poorly crafted mass emails of the past. Today's attacks are targeted, AI-enhanced, and delivered across multiple channels at once.
Classic email phishing
Mass emails with obvious errors and generic subjects. As awareness of basic cues improved, attackers adapted by making lures contextual and targeted.
Spear phishing & BEC
Highly targeted attacks using social media and corporate data to build legitimacy. Business Email Compromise is now one of the most financially damaging attack classes.
AI-enhanced & multi-channel
Generative AI produces convincing messages at speed. Modern phishing spans email, SMS, voice, collaboration tools, and deepfake media, often in the same campaign.
Modern phishing now routinely extends to:
03Where the numbers sit right now
Phishing trends in 2025 and 2026
Organisations worldwide continue to face significant phishing volumes. In Australia, ASD's ACSC received more than 84,700 cybercrime reports in FY2024 to 25, roughly one every six minutes, with phishing, compromised accounts, and identity-information gathering among the top three techniques observed across both government and non-government incidents.
jump in AI-generated phishing late in 2025, peaking at 56% of reported attacks.
Hoxhunt Phishing Trends Report, 2026average cost of a data breach that started with phishing, the most common way in.
IBM Cost of a Data Breach Report, 2025cybercrime reports received by ASD's ACSC in FY2024 to 25, one every six minutes.
ASD's ACSC Annual Cyber Threat Report 2024 to 25Internal-themed lures, messages posing as HR or IT, regularly top the lists of most-clicked simulated phishing subjects, because familiarity breeds trust. Campaigns increasingly use QR codes and redirect chains to obscure malicious URLs and slip past MFA controls.
04Why the click still happens
The human factor and risk exposure
Phishing remains widely successful because the people being targeted still find it hard to distinguish malicious from legitimate communication, even experienced, digital-native users, particularly when attackers use AI-generated text and contextually accurate detail.
Verizon's 2025 Data Breach Investigations Report found phishing was the way in for 16% of breaches, and the human element was involved in 60%, a reminder that awareness alone, without reinforcement, isn't closing the gap.
05Beyond the inbox
The real-world impact of phishing attacks
Phishing is not a theoretical risk. It directly contributes to data breaches, credential theft, ransomware, and BEC losses. Credential theft, frequently initiated via phishing, remains one of the fastest-growing categories of breach in 2025 and 2026 data.
In some recent large-scale incidents, phishing was the entry point to data compromise affecting millions of customer records.
Economically, an attack that starts with phishing can result in loss of sensitive data and intellectual property, direct financial fraud, disrupted operations, and regulatory or reputational damage, often all at once.
06What actually reduces risk
Phishing simulations and organisational resilience
A growing body of evidence shows that well-executed phishing simulation programs meaningfully improve organisational resilience, not as a compliance exercise, but as a measurable behaviour-change tool.
fewer malicious clicks reported by organisations using behaviour-based training.
Hoxhunt, vendor dataimprovement in reporting of real attacks within six months.
Hoxhunt, vendor datadrop in phish-prone rate after 12 months of sustained training.
KnowBe4, 2025The programs that produce these outcomes share four traits: realistic scenarios that mirror current threat patterns, immediate feedback explaining why a lure was malicious, targeted follow-up training based on results, and continuous measurement and refinement. Organisations that treat this as iterative, rather than a once-a-year event, see measurable gains in both detection and reporting, shifting the culture from compliance-driven to behaviour-driven.
07Where this is heading
Outlook and strategic imperatives for 2026
As phishing tactics keep evolving, organisations need layered, dynamic strategies rather than a single annual training event:
AI-augmented phishing and multi-vector campaigns are very likely to keep accelerating through the rest of the decade, which makes human awareness and adaptive training more critical to organisational defence, not less.
·Key takeaways
What this means for your organisation
No amount of technical investment removes the need to train your people: phishing exploits psychology, not just technology.
More than 80% of phishing emails analysed show signs of AI assistance. Generic awareness training is no longer enough on its own.
Scenarios built from live threat intelligence produce measurably better outcomes than recycled templates.
One-off training produces one-off results. In KnowBe4's data, a year of sustained training cut the phish-prone rate from 33.1% to 4.1%.


