SAF's Digital and Intelligence Service launches new training command to keep pace with evolving threats

The new command is a training organisation within the SAF’s Digital and Intelligence Service, or DIS. Its stated purpose is to teach personnel how to extract intelligence using artificial intelligence. That matters because modern intelligence work involves more information and faster-changing threats than traditional methods can easily handle.

A practical example is an analyst using AI to search, sort, translate, or summarise large collections of information before examining the most relevant material. The system can help identify patterns and connections, while trained personnel remain responsible for judging whether the results are reliable and useful. The supplied article does not describe the command’s curriculum or tools in detail.

The launch shows that the SAF is treating training as part of its response to evolving threats. AI may speed up analysis, but it does not replace human judgment. Personnel still need to understand context, check sources, recognise errors, and turn machine-assisted findings into assessments that commanders can use.

Singapore’s Digital and Intelligence Service, or DIS, is a service within the Singapore Armed Forces that combines digital capabilities with intelligence work. The supplied article identifies it as the organisation launching a training command and operating an Open Source Unit. In practical terms, such a service helps the military understand situations and support decisions.

Its work can include gathering information, analysing data, protecting digital systems, and producing intelligence for commanders. For example, personnel may examine publicly available material, compare it with other reporting, and identify developments relevant to defence planning. The article specifically highlights training to extract intelligence using AI, but it does not provide a full list of the DIS’s responsibilities.

The DIS therefore connects information work with military action. Its importance is growing as threats change and digital information expands. A capable service can help commanders make more informed decisions, while its new training command is intended to ensure personnel develop the skills needed for AI-assisted intelligence work.

Open-source intelligence, commonly called OSINT, is intelligence produced from publicly available information. Sources can include news reports, government publications, company records, websites, social media, maps, photographs, and videos. The article points to the DIS’s Open Source Unit and says personnel are being taught to extract intelligence using AI.

AI can help by processing far more text, images, audio, and video than one person can review manually. It can find repeated names or locations, translate foreign-language material, group related reports, flag changes, and rank items for closer examination. Analysts must still verify sources, detect manipulation, and add context. AI output is a lead, not automatically a fact.

OSINT is valuable because public information can reveal developments quickly, but it can also be incomplete, misleading, or deliberately false. The new training command suggests Singapore wants personnel who can use AI efficiently while applying professional judgment. The article does not specify which AI systems or public sources the unit uses.

The supplied article does not state how much information intelligence personnel must examine. It gives no number, volume estimate, or list of source types. What it does make clear is that the SAF is responding to evolving threats by creating training focused on extracting intelligence with AI.

In modern intelligence work, information can arrive continuously through news, websites, social platforms, imagery, documents, and other digital channels. Much of it may be duplicated, irrelevant, poorly organised, or written in different languages. Human analysts can read deeply and understand context, but they have limited time and attention. Reviewing everything manually can delay useful findings.

AI can reduce that burden by searching large collections, sorting material, detecting similarities, summarising content, and highlighting possible changes or links. These tools still require human checks because algorithms can miss context or repeat false claims. The training command’s purpose is therefore not simply speed; it is to build personnel who can combine machine processing with careful analysis.

An intelligence service that cannot keep pace may struggle to identify important developments before they affect Singapore’s security. Information could arrive too slowly, warning signs could be buried, and assessments could be based on incomplete evidence. The supplied article does not describe a specific failure or predict a particular attack; it links the new command to the need to address evolving threats.

For example, if analysts cannot process a fast-moving stream of reports, they may overlook a change in an adversary’s activity or misunderstand its significance. Commanders could then face delayed warnings, fewer options, or decisions made with an incomplete picture. AI can help sort and prioritise information, but human analysts must check accuracy and explain uncertainty.

The broader consequence is reduced decision advantage. Intelligence supports planning, readiness, and responses, so gaps can affect both security protection and military choices. The new training command indicates that the SAF is preparing personnel to use AI more effectively. Its success will depend on skills, reliable sources, and strong human oversight.

Public information is only one part of intelligence work. Militaries can also use human intelligence from people, signals intelligence from communications or electronic activity, imagery and geospatial intelligence from photographs and maps, and measurement or signature data from technical sensors. Cyber intelligence and information shared by partners may provide additional insight.

Each source answers different questions. A public report might reveal what is being discussed openly. Imagery could show a physical change. Signals or technical data might indicate activity, while human reporting could provide context or intent. Analysts compare these streams, test whether they support one another, and identify contradictions rather than treating any single source as complete.

The supplied article specifically mentions the DIS’s Open Source Unit and AI training, but it does not list the other intelligence sources used by Singapore. These alternatives are established components of intelligence practice, not details stated in the article. AI can help combine and search multiple streams, while trained personnel remain responsible for validation, context, and judgement.

Intelligence begins with a question, such as whether a development could affect security. Personnel collect relevant information, including material from public sources, and assess its reliability. They then organise the evidence, compare reports, identify patterns, and separate confirmed facts from assumptions or gaps. This process turns raw information into an assessment.

For example, several reports may describe activity at a location, while imagery shows a physical change there. Analysts can compare dates, sources, and context, then judge whether the evidence supports a meaningful conclusion. AI may search, classify, translate, or summarise the material. Analysts must still check the machine’s output and explain competing interpretations.

The assessment is then communicated in a form commanders can use. It should state the key finding, supporting evidence, uncertainty, and possible implications. The article highlights AI extraction training but does not describe the DIS’s full intelligence process. In general, the value comes from disciplined reasoning, not from collecting information alone.