Southeast Asia’s smart-meter investments are creating a new opportunity for utilities: turning billions of granular consumption readings into operational intelligence. At Enlit Asia 2026 in Jakarta, energy AI company Bidgely will demonstrate how machine learning and big-data analytics can transform advanced metering infrastructure (AMI) data into tools for customer service, revenue protection and distribution-grid planning. The company’s UtilityAI platform will focus particularly on high-bill prediction, non-technical loss detection and visibility into distributed energy resources such as rooftop solar and electric vehicles.
For utilities, installing smart meters is only the beginning of digital transformation.
The more difficult question is what happens after millions of meters begin producing interval-level electricity data.
That question will be central to Bidgely’s presence at Enlit Asia 2026, scheduled for September 22–24 at the Indonesia Convention Exhibition in BSD City, Jakarta. The event brings together utilities, technology providers, regulators and other participants across the ASEAN energy ecosystem, with grid modernization and digital technologies among its major themes.
Bidgely is positioning its UtilityAI platform around that data-to-decision problem. Rather than treating AMI primarily as a more automated way to collect meter readings, the company wants utilities to use consumption data as an analytical layer for understanding customers, detecting suspicious usage patterns and planning distribution assets.
That reflects a broader change in utility technology.
Smart meters are becoming an AI data layer
AMI gives utilities substantially more information than traditional periodic meter readings. Instead of seeing consumption only at billing intervals, utilities can analyze patterns over time and compare changes in usage against customer behavior, equipment and grid conditions.
The challenge is turning that volume of data into actionable information.
Bidgely says its AI and machine-learning models can analyze interval data to generate appliance-level insights, identify abnormal consumption patterns and provide behind-the-meter visibility that can support distribution planning.
The distinction matters because utilities increasingly need to make decisions closer to real time.
Solar generation can change the direction and timing of power flows. Electric-vehicle charging can introduce new peaks. Heat pumps, batteries and other distributed resources can alter load profiles.
A smart meter can measure those changes.
An AI analytics layer can potentially help utilities understand why they are happening and what they mean for the network.
Gartner’s 2026 research on the digital grid describes AI and digital technologies as increasingly important for sensing, balancing and optimizing increasingly complex grids, with the longer-term objective of more adaptive and autonomous grid operations.
From bill shock to proactive customer service
Bidgely’s first focus at Enlit Asia is the customer.
High electricity bills are a particularly visible consequence of changing consumption patterns. Customers may know that their bill increased without knowing which appliances, behaviors or usage periods caused the change.
Bidgely’s approach is to use AI-based disaggregation to provide more granular information about household energy consumption.
That can give consumers appliance-level context while also providing customer-service representatives with additional information when handling billing inquiries.
The potential benefit is operational as much as experiential.
Instead of a service representative working from a total bill and a historical account record, an AI system can surface relevant consumption patterns and anomalies during the interaction. That could help utilities move from explaining a bill after the fact toward identifying unusual consumption before it becomes a customer complaint.
This is increasingly important as electrification makes residential demand more complex.
The utility-customer relationship is shifting from a largely transactional billing model toward a data-enabled energy-management relationship.
AI targets a persistent utility problem: non-technical losses
The second major application is revenue protection.
Non-technical losses—including electricity theft, meter tampering and other discrepancies that are not caused by physical transmission or distribution losses—remain a significant challenge for utilities.
AMI provides more data for detecting suspicious behavior, but simply collecting more data does not automatically identify theft.
Bidgely says its anomaly-detection algorithms can analyze consumption patterns to identify potential cases and generate targeted leads for investigation.
The broader technology principle is important: AI can shift revenue protection from blanket inspection toward risk-based investigation.
Instead of manually reviewing large numbers of accounts, utilities can prioritize customers or locations whose consumption patterns exhibit characteristics associated with potential anomalies.
There is already evidence that AMI can support this type of operational use in Southeast Asia. Thailand’s Metropolitan Electricity Authority, for example, describes AMI-based analytics for tamper detection and non-technical-loss reduction, alongside applications for outage management, demand response and monitoring voltage impacts from rooftop solar.
AI therefore sits on top of an infrastructure layer that utilities are already building.
The opportunity is to make that infrastructure more predictive.
The grid is becoming harder to manage from the top down
The most strategically significant part of Bidgely’s Enlit Asia proposition may be its focus on distributed energy resources (DERs).
Traditional distribution planning could largely work from relatively predictable demand patterns. The growth of rooftop solar, EVs, batteries and other distributed assets makes that assumption less reliable.
A transformer serving a neighborhood with several EVs and solar installations can experience a very different load profile from one serving an otherwise similar area.
Bidgely’s DER Grid Planning technology uses AI-based disaggregation to construct a more detailed picture of behind-the-meter energy use. The company says this can give utilities visibility into transformer capacity, asset stress and opportunities for load shifting.
That changes the planning question.
Instead of asking only where electricity is being consumed, utilities can begin asking what devices are driving that consumption, when they operate and how those loads interact with local grid assets.
That information could help utilities prioritize upgrades rather than automatically expanding infrastructure everywhere.
The potential business implication is significant: better visibility can support more targeted capital expenditure and more efficient use of existing assets.
AI adoption is moving into utility IT and OT
This is part of a wider shift in the energy industry.
Gartner’s 2026 energy and utilities research says organizations are under pressure to use AI, analytics and automation to improve operational performance while addressing regulation, risk and rising demand.
Another Gartner assessment highlights the need for unified data architectures capable of supporting increasingly autonomous decisions, while identifying legacy infrastructure and talent shortages as major risks for energy and utilities organizations.
That makes utility AI fundamentally different from deploying a consumer-facing chatbot.
AI models have to operate alongside systems responsible for billing, customer service, asset management, outage response and grid operations. Data quality, cybersecurity, governance and integration can therefore determine whether an AI deployment produces measurable value.
The cybersecurity dimension is becoming more urgent as well. Recent reporting has highlighted how increased connectivity across energy infrastructure expands the attack surface for utilities and creates new risks around operational technology.
For utilities, AI adoption consequently has to fit within existing enterprise and operational technology architectures rather than operating as an isolated analytics experiment.
Bidgely emphasizes deployment flexibility
That integration challenge is reflected in Bidgely’s deployment strategy.
The company says UtilityAI can operate as a fully managed software-as-a-service platform or within a utility’s preferred cloud and data environment, including Amazon Web Services, Microsoft Azure, Snowflake and Databricks.
The approach is significant because utilities have different requirements around data residency, security, governance and existing infrastructure.
A utility that has already invested heavily in a particular cloud or data platform may be reluctant to create a separate AI data silo.
By supporting established cloud and data environments, Bidgely is effectively positioning UtilityAI as an analytical layer that can fit into existing utility data architectures.
That is increasingly important as AI projects move from pilots toward production.
Enlit Asia reflects the region’s grid modernization push
Bidgely’s appearance also aligns with the broader agenda at Enlit Asia.
The 2026 event is expected to bring more than 11,000 attendees, with more than 2,000 utilities and independent power producers represented, according to event organizers. Its agenda spans grid modernization, AI, data centers, renewable energy and other technologies shaping the region’s electricity system.
Bidgely will participate in sessions covering digitalization and AI, modern power systems, and distribution networks operating under changing demand patterns.
Those themes point to the central issue facing utility AI: the technology has to move beyond dashboards and demonstrations and become part of everyday operational decision-making.
The next phase is turning data into decisions
The smart-meter era created a new source of granular utility data. The AI era is now creating tools for interpreting it.
Bidgely’s strategy illustrates where that intersection is heading.
For consumers, AI can provide greater visibility into energy use. For utilities, the same data can help prioritize revenue-protection investigations. At the grid level, behind-the-meter intelligence can potentially improve understanding of transformer loading, DER adoption and demand flexibility.
The underlying technology is not simply another analytics dashboard.
It is an attempt to create a bottom-up intelligence layer for the distribution grid, connecting what happens inside homes and businesses with decisions made by utility operators.
That distinction will become increasingly important as Southeast Asia adds renewable generation, EVs, batteries and other flexible loads.
The utilities that extract the most value from their AMI investments may not be those with the most meters, but those that can turn the resulting data into reliable operational decisions.
Market Landscape
The utility AI market is expanding across several overlapping categories: AMI analytics, non-technical-loss detection, customer intelligence, DER management, grid analytics, asset optimization and AI-enabled utility operations.
Bidgely competes in an ecosystem that includes established grid and utility technology providers such as Schneider Electric, Oracle, GE Vernova and Siemens, alongside specialist analytics and grid-edge companies.
Enlit Asia’s 2026 exhibitor list illustrates the breadth of this market, with companies spanning utilities, grid technology, AI, distribution, transmission and IT/ICT. Bidgely is listed alongside organizations including GE Vernova, Oracle and Schneider Electric.
The strategic battleground is increasingly shifting from collecting utility data to operationalizing it.
As DER penetration rises, utilities will need better visibility into assets and consumption below the substation level. AI can help bridge that gap, but its value will depend on data quality, system integration, governance and the ability to translate predictions into operational actions.
Top Insights
- Bidgely is taking utility AI beyond smart-meter monitoring, using AMI data for customer intelligence, revenue protection and distribution-grid planning.
- Non-technical-loss detection is becoming more data-driven, with AI identifying anomalous consumption patterns so utilities can focus investigations where risk is highest.
- DER growth is changing distribution planning, making behind-the-meter visibility increasingly important for transformer capacity, asset stress and demand flexibility.
- Cloud flexibility is becoming an enterprise requirement, with UtilityAI supporting deployment across SaaS and ecosystems including AWS, Azure, Snowflake and Databricks.
- Southeast Asia’s AMI investments are creating an AI opportunity, as utilities move from collecting granular consumption data toward using it for operational decisions.
Power Tomorrow’s Intelligence — Build It with TechEdgeAI
