floLIVE

03/11/2026

271 point-of-sale terminals. One week in October. 1,508 command-and-control alerts, all pointing to the same server.

The terminals shipped with a factory-installed Android app linked to a known malware family. No one on the customer’s side installed it.

FLOLIVE® network-level security monitoring flagged the traffic. Our team traced it through the alert history, packet captures, and a comparison across the fleet, without installing an agent on the devices.

The terminals were not under active attack during the analysis. Of the 1,508 connections, 1,497 went unanswered and the remaining 11 received empty replies. A quiet server does not make the devices safe.

The immediate containment step is to block the malicious hostnames on the customer’s sub-account APNs while the fleet is remediated with clean firmware.

Ahead of peak retail season, network visibility can expose traffic that a device does not report.
Link in the comments.

02/11/2026

Add a third carrier to your IoT offer and a new question shows up on day one.
Who sees the policies, the diagnostics, and the customer experience across all three?

That's the conversation we are bringing to MVNO Nation Europe in Alicante, November 3 to 5.
FLOLIVE® works with MVNOs and connectivity providers on managing several carriers from one platform, with local routing and visibility across markets.

Have a coffee on us during the coffee break and come find us at booth #16. Bring a live requirement: a market you want to enter, a carrier you are adding, a customer asking for local routing. We'll work through it with you.

29/10/2026

Where an enterprise runs AI inference has direct implications for the network connecting its devices.

A robot running models locally needs connectivity for model updates. A device relying on inference in the cloud has different requirements for latency and resilience. Those requirements can change as the deployment evolves.

On the IoT & AI Leaders Podcast, FLOLIVE® CEO Nir Shalom joined Nick Earle to share his view of how networks need to evolve around these enterprise needs.

His central argument: enterprises should be able to define what their applications need and have the network adapt accordingly. That means greater control over routing and network behavior, with the potential for AI agents to request and release capabilities within enterprise-defined policies.

For businesses building physical AI, connectivity needs to be part of the architecture conversation from the start.

Thank you to Nick and the Eseye team for hosting the discussion.

Watch the episode: Physical AI Needs a New Network https://hubs.li/Q04CG3zy0

27/10/2026

We’re honored to receive Frost & Sullivan’s 2026 Technology Innovation Leadership Award for Smart Connectivity in Asia Pacific.

The recognition reflects our work at FLOLIVE® to help businesses scale IoT across APAC while meeting local performance, data-routing, and cost requirements.

Frost & Sullivan’s assessment highlighted our distributed core architecture and in-country local breakout, which bring traffic exits closer to devices and inference locations. This supports latency-sensitive IoT and AI applications and gives businesses greater control over where their data travels.

The assessment also recognized how we make network telemetry available through APIs, giving customers network context they can use in their own analytics and AI systems.
Thank you to Frost & Sullivan for this recognition, and to our customers, partners, and team for the work behind it.

Read the full announcement: https://hubs.li/Q04CmQ6G0

26/10/2026

A couple of seconds of network downtime can affect how a real-time AI application performs, and possibly even its compliance requirements.

Steffen Sorrell, Chief of Research at Kaleido Intelligence, explains why in this clip. IoT devices feeding real-time AI are no longer passive data collectors. Their data can drive actions as they happen, which makes the application sensitive to jitter, latency, and downtime.

The full webinar with Kaleido Intelligence and FLOLIVE® looks at what that means for connectivity infrastructure, and what enterprises should expect from their providers.

Webinar recording and the white paper that goes with it: https://hubs.li/Q04Cf-yb0
Registrants receive both.

14/10/2026

Connectivity is part of AI data quality. When IoT devices lose coverage, drop packets, hit congestion or switch networks, the data your AI depends on arrives late or incomplete.

That's the case Transforma Insights makes in its new position paper, "Connecting AI: The critical role of connectivity in AI data quality."

The damage shows up as missed events and false alerts. A routing AI keeps planning around a van's position from 15 minutes ago. And since people often can't trace the problem back to the network, they blame the model.

On Thursday, October 29, Matt Hatton of Transforma Insights and Assaf Harel of FLOLIVE® will walk through how connectivity shapes the data your AI runs on, and what AI-ready connectivity looks like.

8:00 PT / 11:00 ET / 15:00 UK / 16:00 CET. Virtual.
https://hubs.li/Q04z-Xy00

12/10/2026

Your AI flagged an event that never happened.

The connection dropped for a few minutes. The model kept running.

When a device loses coverage, AI can continue working with incomplete or delayed data. The output looks plausible, but it can be wrong. Teams can spend weeks tuning the model while the underlying problem sits in the network.

A Transforma Insights position paper, sponsored by FLOLIVE®, traces that chain from connectivity to business outcome: how coverage gaps and packet loss can become missed events and false alerts.

It also distinguishes three types of AI using IoT data: customer AI, real-time operational AI and security AI. One device can feed all three, but each has different data requirements.

Before tuning the model again, check whether its inputs arrived complete and on time. Better AI starts at the network layer.

08/10/2026

A tracker crosses a border and attaches to the wrong network.

The roaming bill climbs. The signal drops. Nobody on your team touched anything.

IoT devices have no screen and no user to pick a better network. Steering makes that choice for them, using rules you set for signal strength, cost, and regulatory compliance.

FLOLIVE® runs steering at three layers inside one platform: SIM-based, core network, and billing-based. You configure the rules in the Connectivity Management Platform, and devices switch networks dynamically.

The result: better performance, lower cost at scale, and fewer manual fixes.

Your devices choose the right network on their own, so your team can focus on what those devices do for your business.

07/10/2026

Extending IoT coverage in Brazil should also make connectivity easier to manage.

For farms, fleets and remote utility sites, cellular coverage gaps can mean missing data and reduced visibility into operations. Adding satellite brings greater reach. Bringing both networks into one service makes that reach practical to manage at scale.

Last night at Futurecom, our team announced announced FLOLIVE® and Skylo are combining local cellular connectivity in Brazil with satellite NTN coverage where cellular signal is poor or unavailable.

Devices use cellular where coverage exists, with satellite extending connectivity beyond the terrestrial footprint. Both are managed through the Flolive platform, with one invoice.

Through our Brazilian partners iuh! Digital and NLT, this gives enterprises a way to extend IoT deployments into harder-to-reach locations while keeping connectivity operations in one place.
Explore the service and talk to our Brazil team:
https://hubs.li/Q04zsymr0

05/10/2026

By 2030, 8 in 10 frontline workers at major companies will have an AI system working next to them, according to recent Gartner research.

Most of those systems won't be typing prompts. They'll be picking things up, driving, or flipping a switch.

Gartner's read: IoT infrastructure stops being a way to collect data. It becomes the layer that carries out the decision.

An autonomous agent switching off a valve because it misread a sensor is a different kind of failure than a bad data point. Gartner's answer: give every agent an identity, permissions, and an audit trail, same as a person.

If an agent on your network took the wrong action tomorrow, could you show who requested it and why?

Read more on IoT Business News:

Gartner predicts physical AI and autonomous agents will transform connected systems by 2030, driving new IoT infrastructure roles, energy management, and AI governance in enterprises.