noworkleft.com

Home › Jobs › Information Security Analysts

Computer and Mathematical · SOC 15-1212

Information Security Analysts

Plan, implement, upgrade, or monitor security measures for the protection of computer networks and information. Assess system vulnerabilities for security risks and propose and implement risk mitigation strategies. May ensure appropriate security controls are in place that will safeguard digital files and vital electronic infrastructure. May respond to computer security breaches and viruses.

Will AI replace this job?

The data cannot say, and anyone who gives you a percentage for “replaced” is guessing. What it does show is a gap between three things, for the 11 tasks O*NET lists for this occupation:

Could an LLM do it?

9%–36%

of tasks: an LLM alone could do the task or cut its time by half. Human raters say 9%, GPT-4 says 36%.

Counting tasks that need extra software around the model: 91%–100%.

Is it happening?

5 of 11 tasks

show measurable use in Claude conversations (Anthropic Economic Index). Anthropic’s occupation score, which counts only tasks that were also rated feasible, is 49%. Observed is a subset of “could” by construction.

Claude only, mostly US users, sampled periods in 2025. Not all AI use, and not employment.

What is left?

0 of 11 tasks

were rated “no LLM speed-up” by every rater (so, by Anthropic’s method, they cannot show up as observed use either). That is the part of the job these studies see no sign of AI touching.

It does not mean those tasks are safe. It means these methods found nothing there.

Task by task

Sorted with the tasks seen in use first. The three coloured squares are the three raters, in this order: human raters, GPT-4, GPT-4 with an alternative rubric. “Usage score” is Anthropic’s published task_penetration value; 0 means none seen.

LLM aloneneeds extra softwareno speed-up
Task (O*NET 31.0)Could an LLM?Claude usage score
Perform risk assessments and execute tests of data processing system to ensure functioning of data processing activities and security measures. all raters 0.95
Modify computer security files to incorporate new software, correct errors, or change individual access status. all raters 0.95
Document computer security and emergency measures policies, procedures, and tests. all raters 0.83
Monitor use of data files and regulate access to safeguard information in computer files. all raters 0.97
Coordinate implementation of computer system plan with establishment personnel and outside vendors. all raters 0.88
Develop plans to safeguard computer files against accidental or unauthorized modification, destruction, or disclosure and to meet emergency data processing needs. all raters none seen
Monitor current reports of computer viruses to determine when to update virus protection systems. all raters none seen
Encrypt data transmissions and erect firewalls to conceal confidential information as it is being transmitted and to keep out tainted digital transfers. all raters none seen
Review violations of computer security procedures and discuss procedures with violators to ensure violations are not repeated. all raters none seen
Train users and promote security awareness to ensure system security and to improve server and network efficiency. all raters none seen
Confer with users to discuss issues such as computer data access needs, security violations, and programming changes. 2 of 3 raters none seen
How to read this. These are averages over task lists written for the US labour market, rated by people and by GPT-4 using early-2023 model capabilities, next to a count of what one company’s chatbot users actually ask for. A task being “exposed” says nothing about whether employers will cut jobs, shift work, or hire more of you. See how to read an exposure score.

Try it with your own task list

Sources: task list from O*NET® Database (U.S. Department of Labor, CC BY 4.0; modified by noworkleft.com); “could” labels from Eloundou et al., GPTs are GPTs (OpenAI/University of Pennsylvania, MIT licence, labels from early-2023 models); “seen in use” from the Anthropic Economic Index (Claude usage only, CC BY). Dates, versions and limits.

Other jobs in Computer and Mathematical

Data built 2026-11-06. O*NET 31.0. Task match rates: 99.4% with the OpenAI/Penn labels, 99.2% with Anthropic usage.