Many organizations have begun to embrace the use of Artificial Intelligence (AI) in developing their companies, both in business and in education. One of the AIs used by these organizations is ChatGPT. ChatGPT, or Chat Generative Pre-trained Transformer , is a language tool or model developed specifically to converse like a human. ChatGPT provides all the information needed by its users from the data it has gathered and collected from the internet.
So, how can ChatGPT assist in company development?
- Assisting in Content or Material Creation
ChatGPT is used to help companies in content creation, such as providing variations and ideas that align with the company's image. This includes everything from catchy titles to outlines of content that will later be published on social media. In the field of education, ChatGPT can assist in material creation, from research to curriculum development in subjects.
- Consumer Service
The presence of ChatGPT makes it easier for companies to provide consumer services. ChatGPT's ability to respond like a human can help companies enhance consumer services with quick response capabilities. It can tailor responses according to the needs and preferences of consumers.
- General Assistance to Companies
ChatGPT can provide services such as creating presentations, translation services, and assisting in employee selection by providing necessary questions for the company.
To maximize the benefits of ChatGPT, understanding how to provide commands or prompts is very important. This knowledge allows users to obtain accurate information or responses according to their needs. For those of you who wish to delve deeper, there are many learning resources available, including classes and training on various educational platforms.
Since this article was written, the two uses mentioned above already have written regulations, no longer just work ethics. The European Union has enacted Regulation (EU) 2024/1689, commonly known as the AI Act. This regulation takes effect on August 2, 2026, and the European Commission, along with member state authorities, will begin enforcing it on the same date. This is EU law, not Indonesian law, so it does not automatically bind companies here. However, if your company serves clients, candidates, or users in Europe, this regulation applies to you, and even if it does not, the list of obligations is the most detailed reference currently available.
The first use is consumer services. From August 2, 2026, transparency rules require chatbots and other interactive AI systems to inform their users that they are interacting with a machine, not a human. Content generated or modified by AI must also be recognizable, including through machine-readable tags, and deepfake must be labeled. Thus, the idea of "responding like a human" mentioned in this article has an explicit limit: it may sound human-like, but it must not impersonate a human.
The second use, employee selection, is considered much riskier. The AI Act classifies AI systems for recruitment and workforce management as high-risk, with examples provided by the Commission itself including CV sorting software. The obligations are significant: assessment and risk mitigation, quality data to minimize discriminatory outcomes, activity logging for traceability, technical documentation, adequate information for users, human oversight, and resilience and accuracy. The effective date was changed: through a simplification package called AI Omnibus, effective July 27, 2026, the obligations for high-risk systems in Annex III, including in the employment sector, were postponed to December 2, 2027. One thing that has already been prohibited since February 2025 is emotion recognition in workplaces and educational institutions.
What you can do now does not wait until 2027 and does not wait for Indonesian regulations. Note where AI is used that touches people, not just documents. For chatbots, state from the first greeting that it is a machine, and provide a way to a human. For employee selection, keep the reasons for decisions, not just the scores, and ensure there is a human who can actually overturn the results, not just press an agree button. These three notes are inexpensive to implement now and costly to create later, regardless of the jurisdiction that ultimately applies to you.
Sources
- European Commission – AI Act, official policy page, includes risk levels, examples of recruitment as high-risk, and implementation timelines (last updated August 3, 2026)
- European Commission, press release July 31, 2026 – Commission starts enforcing AI Act rules and new transparency requirements on 2 August, chatbot obligations to identify as AI and content labeling
- European Commission, July 27, 2026 – AI Omnibus enters into force, postponement of Annex III obligations to December 2, 2027, and Annex I to August 2, 2028