BENGALURU, Oct 12: Indian technology services company HCLTech has entered into a strategic partnership with artificial intelligence developer Anthropic to expand the use of agentic AI across businesses, with a particular emphasis on cybersecurity, safety and responsible deployment.
Under the collaboration announced on Monday, HCLTech will establish a dedicated business unit and a centre of excellence to develop and implement AI solutions for enterprise customers. The initiative is aimed at helping organisations move beyond conventional AI assistants towards systems capable of carrying out more complex, multi-step tasks.
The agreement brings together HCLTech’s experience in enterprise technology services with Anthropic’s AI products, including Claude, Cowork and Claude Code. The companies intend to develop industry specific applications that can be integrated into existing business operations.
The partnership reflects a broader shift in the technology sector as companies explore how AI can be incorporated into software development, IT support, business process management and other activities that traditionally require substantial human involvement.
According to Reuters, the collaboration is designed to accelerate enterprise adoption of agentic AI while prioritising security and governance.
Focus shifts towards AI agents
Traditional AI tools are commonly used to answer questions, summarise documents, generate content or assist employees with individual tasks. Agentic AI systems are designed to take a more active role by planning workflows, using digital tools and completing several connected actions towards a defined objective.
For example, an AI agent supporting an IT department could help identify a technical problem, examine relevant system information, suggest a solution and prepare the necessary steps for remediation. Depending on the permissions granted, some actions could be automated while others would require approval from an employee.
Such capabilities could reduce the time spent on repetitive work and allow staff to concentrate on complex decisions, customer requirements and strategic responsibilities.
However, the transition also introduces new challenges. An AI system that can interact with business applications or access internal information requires carefully defined permissions, reliable monitoring and safeguards against unauthorised activity.
Companies must also ensure that automated decisions can be reviewed and that responsibility remains clear when an AI agent produces an incorrect result.
These considerations are particularly important for organisations operating in sectors such as banking, healthcare, telecommunications and manufacturing, where mistakes involving confidential information or critical systems could have serious consequences.
Dedicated business unit to support implementation
HCLTech’s planned business unit will focus on developing enterprise applications based on agentic AI. The centre of excellence is expected to support the development and implementation of solutions suited to the operational requirements of different industries.
Instead of treating AI adoption as a stand-alone software installation, businesses increasingly need assistance with integrating new tools into their existing technology infrastructure. This can involve connecting applications, reviewing data access arrangements, redesigning workflows and training employees to work alongside automated systems.
The partnership could help customers address these requirements through a combination of AI products, consulting, engineering and implementation services.
For large organisations, successful deployment often depends on whether an AI solution can operate reliably within established processes. A demonstration may show that an agent can complete a task, but a production environment also requires consistent performance, secure access, auditability and procedures for handling failures.
HCLTech’s role will therefore extend to helping enterprises determine where automation is appropriate and how it can be introduced without disrupting essential operations.
The companies have not publicly detailed every planned application or provided a comprehensive timetable for rolling out the partnership’s solutions.
AI development and software engineering
The inclusion of Claude Code in the collaboration points to software engineering as one area of potential application.
AI coding tools can assist developers in understanding unfamiliar code, generating software, identifying errors, preparing tests and suggesting corrections. When incorporated into established development processes, these capabilities may help engineering teams work more efficiently.
Nevertheless, AI-generated code must still be tested and reviewed. A system may produce code that appears functional but contains security weaknesses, performance problems or errors that emerge only under particular conditions.
Enterprise customers must also consider intellectual property, data confidentiality and the handling of proprietary source code when selecting AI development tools.
A carefully managed implementation can combine automated assistance with established software testing, code review and approval procedures. This approach allows organisations to benefit from faster development while preserving accountability for the final product.
Cybersecurity remains a central concern
As AI systems gain the ability to interact with enterprise applications, cybersecurity becomes a fundamental requirement rather than an optional feature.
An AI agent may need access to documents, customer records, development environments or internal management systems to complete assigned work. If those permissions are excessive, an error or malicious instruction could expose information or trigger an unintended action.
Businesses therefore need to establish clear boundaries around the information agents can access and the operations they can perform. Sensitive activities may require additional verification or explicit human authorisation.
Organisations must also monitor agent activity, maintain records of important actions and establish procedures for suspending automated processes when suspicious behaviour is detected.
The partnership’s stated emphasis on safety, cybersecurity and governance reflects the growing importance of these controls as AI moves into operational business environments.
Effective safeguards must cover the entire deployment process, including model selection, system integration, employee access, testing and ongoing monitoring. Security measures also need to be updated as applications and threats evolve.
Enterprise adoption requires workforce preparation
The expansion of agentic AI is likely to change the way employees interact with enterprise software. Workers may increasingly delegate routine digital tasks to automated systems while retaining responsibility for reviewing results and handling exceptions.
This transition will require training in AI capabilities, limitations and appropriate use. Employees need to understand when a system can be trusted to perform a routine task and when independent verification is necessary.
Businesses will also have to assess which processes are suitable for automation. Tasks involving repetitive data handling or predictable workflows may be easier to automate than activities requiring sensitive judgement, negotiation or interpretation of incomplete information.
Clear internal policies can help employees understand how AI tools should be used and which categories of information must remain restricted.
The long-term benefits will depend not simply on the number of AI tools introduced but on whether they improve productivity, reliability and service quality.
A competitive opportunity for India’s IT sector
The agreement highlights the evolving role of Indian IT services companies in the global AI market. Technology service providers have traditionally helped customers implement software, maintain infrastructure and manage digital transformation projects.
As generative and agentic AI become more widely available, these companies are seeking opportunities to help enterprises deploy the technology securely and integrate it into existing systems.
Demand may increasingly favour providers that can combine AI expertise with industry knowledge, cybersecurity capabilities and long-term operational support.
For HCLTech, the collaboration provides an opportunity to develop additional services around enterprise AI adoption. For Anthropic, working with an established technology services provider could help its products reach organisations that need assistance implementing AI across complex operations.
The partnership’s eventual impact will depend on customer adoption, the effectiveness of its solutions and the measurable benefits businesses obtain from deployment.
Enterprises will need to evaluate whether these tools deliver sufficient improvements to justify implementation costs, training requirements and ongoing security obligations.
The latest agreement underlines an important development in enterprise technology: AI is moving from an experimental productivity tool towards systems that can participate more directly in business workflows. Its success, however, will depend on maintaining security, human accountability and reliable performance as automation expands.