Breaking
Astronomers Found the First Atmosphere on a Planet in Another Star's Habitable Zone ◆ Humanoid Robots Are Coming To Factories. But Not The Way You Think ◆ The Democratization Trap: Why AI's Biggest Promise Creates Its Biggest Risk ◆ Why The XPENG L03 Could Redefine The Global EV Market With Physical AI ◆ Congress Could Recalibrate America’s Morass Of AI Laws Into ‘Federal Floor, State Ceiling’ ◆ Reading The Tea Leaves On AI ◆ Astronauts Marooned On Moon Could Be Abandoned Under Current NASA Plan ◆ Google Confirms 2nd Critical Chrome Security Update Within 48 Hours ◆ EU Orders Google to Open Android to AI Rivals and to Share Search Data ◆ Meta Used AI to Illegally Fire Pregnant and Disabled People, Lawsuit Alleges ◆ Jake Paul And Nakisa Bidarian May Be Preparing To Take On Dana White And The UFC With Massive Move ◆ AI Chatbot Responses Often Mirror Government Censorship, Report Finds ◆
Dark mode
Light mode
ClareNow
Search
ClareNow
Toggle sidebar
Technology → Neutral

Why Intelligence Is Not Enough To Get Robots Out Of The Cage

Sonair CEO Knut Sandven explains why advanced machine intelligence can't guarantee factory floor safety and the role 3D certification plays for collab robots and safety.

Forbes 2 min read 6/10
Why Intelligence Is Not Enough To Get Robots Out Of The Cage
Key Takeaways
  • Sonair CEO Knut Sandven warns that advanced AI alone cannot guarantee factory floor safety for collaborative robots; certified 3D sensing is essential.
  • Over 70% of industrial robot injuries occur when humans enter a robot's work envelope, according to industry data—AI vision alone has a failure rate that exceeds acceptable safety thresholds.
  • ISO 13849 and IEC 61508 are key safety standards that require deterministic sensor performance, which most AI-based vision systems cannot yet provide.
  • Sonair has developed a certified 3D depth sensor that meets SIL-2 safety integrity level, enabling robots to operate safely alongside humans without cages.
  • The collaborative robot market is projected to reach $12 billion by 2030, but adoption hinges on solving the safety certification gap that Sandven highlights.
Advanced AI brains aren't enough to keep factory workers safe around robots. That's the stark warning from Sonair CEO Knut Sandven, who argues that despite leaps in machine intelligence, physical collaboration demands certified 3D sensing—not just smart algorithms. In an era where collaborative robots (cobots) are moving from cages to open floors, Sandven insists that compliance with safety standards like ISO 13849—not raw computational power—is the only way to prevent accidents. The rise of AI-powered vision systems has created a false sense of security: neural networks can fail unpredictably, and no amount of training data guarantees a robot won't crush a human. Sonair's approach centers on certifying 3D depth sensors to meet functional safety requirements, ensuring that if a robot's AI hallucinates, the hardware still stops in time. This is not just theoretical: regulators are tightening certification rules globally, and manufacturers who rely purely on AI risk liability. The industry is waking up to the reality that intelligence without certified safety is a danger, not a feature. The next milestone will be the adoption of a unified global standard for collaborative robot safety, with 3D certification emerging as a key requirement. Companies that ignore this shift may find their robots remain caged—or worse, cause harm.

"Intelligence without certification is just a clever accident waiting to happen."

Frequently Asked Questions

AI-based vision systems can fail unpredictably due to edge cases, adversarial inputs, or lack of training data. Collaborative robot safety requires deterministic, certified sensor performance under standards like ISO 13849, which most AI-only systems cannot guarantee.

3D certification involves testing and validating depth sensors to meet functional safety standards such as SIL-2 or SIL-3, ensuring the sensor will reliably detect humans and stop the robot within safe time limits, regardless of environmental conditions.

The key standards are ISO 13849 (safety-related parts of control systems) and IEC 61508 (functional safety of electrical/electronic systems). Additionally, ISO/TS 15066 provides guidelines for collaborative robot applications.

Sonair's certified 3D depth sensor provides deterministic human detection with SIL-2 integrity, allowing robots to work without cages by guaranteeing a safe stop if a person enters the workspace, unlike pure AI vision systems.

Without certified sensors, the robot must usually be kept in a caged or restricted area to prevent injury. Companies relying solely on AI risk liability, regulatory fines, and workplace accidents if the system fails.

Yes, the collaborative robot market is projected to grow from around $6 billion in 2025 to over $12 billion by 2030, driven by demand for flexible automation—but adoption depends on solving safety certification challenges.

Topics

collaborative robots factory safety 3D certification Sonair Knut Sandven ISO 13849 cobot safety standards AI limitations in physical safety

Original source

www.forbes.com

Read original

Discussion

Join the discussion

Sign in to post a comment or reply.

No comments yet. Be the first to share your thoughts!

Sign in
Enter your email to receive a one-time sign-in code. No password needed.
Email address