Security Journey brings security training management into Claude and ChatGPT
Security Journey now lets programme owners assign courses, chase stragglers and generate reports from inside Claude and ChatGPT, at no extra cost for existing customers.
Why it matters
- Security Journey launched Claude and ChatGPT integrations for training programme management, available now at no additional cost to existing customers.
- The integrations are built on the Model Context Protocol and powered by Security Journey's intelligence layer, Aspen, with Adapt and Assist functions.
- Cited research: one in four workers spend over seven hours a week moving information between applications; workers switch apps about 1,200 times a day.
- OpenAI data cited by Security Journey showed weekly ChatGPT Enterprise messages grew roughly eightfold in 2025.
- CEO Dan Newton: "Engagement is the biggest challenge we hear from customers, and the clearest indicator of whether a training program is working."
Security Journey has launched integrations that let customers manage security training from inside Claude and ChatGPT, without opening a separate dashboard. The move targets a concrete productivity cost: the company cites research showing one in four workers spend more than seven hours a week copying information between applications, reconciling data and feeding context into AI tools.
The integrations are available now to existing Security Journey customers at no additional cost. They let programme owners identify learners who are behind, assign courses, answer programme questions and generate reports — all in natural language, inside the assistant they already have open.
The stakes here are straightforward. Security training only works if people actually complete it, and completion depends on managers catching stragglers early. If the check-in happens where the manager already works, the argument goes, it happens more often.
What can users actually do from inside the assistant?
Within Claude and ChatGPT, programme owners can ask plain-language questions about learner progress and take action in the same interaction. Security Journey gives one concrete example: a manager asks which team members have not started secure coding training, then enrolls or prompts those people without leaving the assistant.
The supported actions, as described by the company:
- Identify learners who are behind on their training
- Assign courses to individuals or teams
- Answer questions about the training programme
- Generate reports on progress
Access follows existing Security Journey permissions. Users can view and manage only the information they are already authorised to see, which matters for training data tied to individual performance.
The offering is free for existing customers, which signals that Security Journey sees this as retention and engagement infrastructure rather than a new revenue line.
How is it built?
The integrations rest on two pieces of technology. The first is the Model Context Protocol, the open standard that lets AI assistants connect to external tools and data sources. The second is Security Journey's own intelligence layer, called Aspen, which sits at the centre of the platform.
Aspen has two main functions, according to the company:
- Adapt uses organisational data to link training more closely to recurring software vulnerabilities. It also generates updated rule files for AI coding tools, delivered as pull requests for human review.
- Assist supports programme administration and learner guidance through natural-language interaction — the engine behind the new Claude and ChatGPT experience.
The Adapt function is worth pausing on. By tying training content to the vulnerabilities an organisation actually sees repeatedly, it attacks a long-standing weakness of security education: generic modules that don't reflect what a specific team's code actually breaks on. Pushing updated rules into AI coding tools as reviewable pull requests keeps a human in the loop.
Why engagement is the problem
Dan Newton, Chief Executive Officer of Security Journey, said customers had made clear that learner participation remains the central issue. "Engagement is the biggest challenge we hear from customers, and the clearest indicator of whether a training program is working," Newton said.
He added: "Bringing Security Journey into Claude and ChatGPT means program owners can engage their people faster and more precisely, without ever leaving their AI assistant."
A learning and development leader at a software company, quoted by Security Journey, put the customer side plainly. "AI has changed what we expect from every tool we use. Claude is increasingly where our team spends its day, so being able to check who hasn't started their training and get them enrolled or nudged without switching tools is a big deal. It simply fits into how we already work," the leader said.
That last line is the whole thesis of the product. The friction isn't the training itself — it's the switching.
How big is the switching problem?
The numbers Security Journey cites to justify the integration add up to a real cost:
- One in four workers spend more than seven hours a week copying information between applications, reconciling data and feeding context into AI tools.
- Harvard Business Review research, cited by the company, found workers switch between applications and websites about 1,200 times a day.
- Those workers lose nearly four hours a week just regaining focus after interruptions.
- OpenAI data, also cited by Security Journey, showed weekly ChatGPT Enterprise messages grew roughly eightfold in 2025.
The last figure explains the timing. If enterprise usage of ChatGPT grew that fast in a single year, the assistant has become daily infrastructure for a large share of knowledge workers. Meeting them there is cheaper than pulling them back to a dedicated dashboard.
Does this fit a wider shift?
The launch reflects a broader move in enterprise software towards embedding AI into existing workflows rather than requiring staff to open dedicated systems for each task. Vendors are testing whether conversational interfaces can become a control point for routine business processes — assigning work, tracking uptake, measuring progress — rather than simply a layer for answering questions.
That test is particularly relevant in security training, where organisations face pressure to improve workforce skills in secure coding and AI use while the tools themselves keep changing. Administrative overhead is often the reason programmes stall.
Security Journey says it is also working to bring a similar experience to learners inside the tools they use every day. For now, though, the immediate focus stays on programme owners who need to track uptake, assign work and measure training progress from within Claude and ChatGPT.
The bet, in plain terms, is that the future dashboard is a conversation. If Security Journey's engagement numbers move, expect more training and compliance vendors to follow the same path through the Model Context Protocol.
Source: itbrief.com.au
More from Sophie Lindqvist
Show full bio
Staff writer covering regulation, policy and their impact on artificial intelligence and machine learning at AI In Context.
749 articles