Radiation Emissions of Popular Smartphones
Smartphones have become an integral part of our everyday lives. From work and school to daily tasks, these handheld devices have brought everything into the palm of our hands.
Most people spend 5-6 hours on their phones each day. And, given that our phones emit a tiny amount of radiation, we’re exposing ourselves to radiation for hours each day.
But different phones emit different amounts of radiation.
With the help of data collected by the German Federal Office of Radiation Protection, we visualize the radiation emissions of some popular smartphones in the market today.
Radiation and SAR Values of Smartphones
Smartphones and other mobile devices emit tiny amounts of radiofrequency (RF) radiation. Humans can absorb this radiation when the smartphone is being used or is lying dormant anywhere near their bodies.
The parameter used to measure phone radiation emissions is the Specific Absorption Rate (SAR). It is the unit of measurement that represents the quantity of electromagnetic energy absorbed by the body when using a mobile device.
The Council of the European Union has set radiation standards for cell phones at 2 watts per kilogram, measured over the 10 grams of tissue that is absorbing the most signal.
SAR values are calculated at the ear (speaking on the phone) and at the body (kept in your pocket). For the purposes of this article, we’ve used the former calculations.
Smartphones With the Highest Levels of Radiation Emissions
The Motorola Edge has the highest radiation emission with a SAR value of 1.79 watts of radiation per kilogram. That’s significantly higher than most other smartphone models in the market today and close to the limits set by the EU for cellphones.
Coming in second is the Axon 11 5G by ZTE with 1.59, followed by the OnePlus 6T at a close third with 1.55 W/kg. The Sony Experia AX2 Plus with 1.41 and the Google Pixel 3 XL and 3A XL at 1.39 round out the top five.
Here is a look at the 10 smartphones that emit the highest level of radiation:
Now that we have detailed the worst offenders let’s look at the smartphones with the lowest levels of radiation emissions.
Smartphones With the Lowest Levels of Radiation Emissions
The smartphone with the lowest SAR value is the ZTE Blade V10, with 0.13 watts of radiation per kilogram.
Mobile devices by Samsung carry some of the least radiation risks. The company has four phones considered to be the best in the category. The Galaxy Note 10+ is the best model in their line-up, emitting a meager 0.19 watts per kilogram.
Here is a look at the 10 smartphones that emit the lowest levels of radiation:
There is currently no significant research proving the harmful effects of phone radiation.
Despite this, people who are in contact with their devices for extended periods can at least quantify their radiation exposure and make choices about which brands serve their needs.
AI
Ranked: The Smartest AI Models of 2026
See the smartest AI models in 2026, ranked by Mensa Norway IQ scores from TrackingAI’s benchmark of leading chatbots and vision models.
Published
19 hours ago
on
April 24, 2026
Ranked: The Smartest AI Models of 2026
See visuals like this from many other data creators on our Voronoi app. Download it for free on iOS or Android and discover incredible data-driven charts from a variety of trusted sources.
Key Takeaways
- Grok-4.20 Expert Mode and OpenAI GPT 5.4 Pro (Vision) tie for the top spot in TrackingAI’s April 2026 Mensa Norway benchmark, each scoring 145.
- The top tier is getting crowded, with several leading models now separated by only a few points.
- Scores have risen sharply from 2025, highlighting how quickly frontier AI reasoning has improved on visual pattern-recognition tests.
The race to build smarter AI models is getting tighter at the top.
This visualization, part of Visual Capitalist’s AI Week, sponsored by Terzo, ranks leading systems using data from TrackingAI, which benchmarks models on the Mensa Norway IQ test as of April 2026.
The results show both who leads today and how little now separates the top contenders, with multiple frontier models clustered near the top of the leaderboard.
A Tie at the Top
The ranking offers a snapshot of how today’s leading AI models perform on abstract pattern-recognition tasks, and just how close the race has become.
As the table below shows, only a small gap now separates the top models:
| Model | Mensa Norway IQ (April 2026) |
|---|---|
| Grok-4.20 Expert Mode | 145 |
| OpenAI GPT 5.4 Pro (Vision) | 145 |
| Gemini 3.1 Pro Preview | 141 |
| OpenAI GPT 5.4 Thinking (Vision) | 139 |
| OpenAI GPT 5.3 | 136 |
| Grok-4.20 Expert Mode (Vision) | 133 |
| OpenAI GPT 5.4 Thinking | 133 |
| Meta Muse Spark | 133 |
| Gemini 3.1 Pro Preview (Vision) | 132 |
| Qwen 3.5 | 130 |
| Claude-4.6 Opus | 130 |
| Kimi K2.5 | 127 |
| Manus | 115 |
| DeepSeek R1 | 112 |
| DeepSeek V3 | 111 |
| Gemini 3.1 Flash Preview | 110 |
| Llama 4 Maverick | 110 |
| OpenAI GPT 5.3 (Vision) | 109 |
| Claude-4.6 Sonnet | 106 |
| Bing Copilot | 101 |
| Perplexity | 97 |
| Mistral Medium 3.1 | 96 |
| Claude-4.6 Sonnet (Vision) | 94 |
| Claude-4.6 Opus (Vision) | 82 |
| Llama 4 Maverick (Vision) | 79 |
| OpenAI GPT 5.4 Pro | 73 |
The biggest takeaway is how compressed the top of the leaderboard has become. Grok-4.20 Expert Mode and OpenAI GPT 5.4 Pro (Vision) are tied for first at 145, while Gemini 3.1 Pro Preview follows closely at 141.
That narrow spread suggests frontier AI models are increasingly converging at the top, where a difference of just a few points can shift the rankings.
The gains from 2025 are also notable. Last year’s top score was 135, compared with 145 in this year’s results, highlighting the speed at which leading models are improving on this benchmark.
Not all models are keeping pace. Among major AI developers, Mistral’s top model ranks lowest in this dataset, scoring 97—well below the leading group.
How TrackingAI Runs the Test
TrackingAI uses the public Mensa Norway test, a set of 35 visual-pattern puzzles. For non-vision models, the questions are verbalized, while vision models receive the original images directly.
As a result, these results are best understood as a benchmark comparison—not a definitive measure of overall intelligence. Because the test is fundamentally visual, model scores can vary depending on how the questions are presented.
Why This Benchmark Matters
TrackingAI’s leaderboard is useful because it offers a simple, familiar way to compare reasoning performance over time. The site also notes that if a model refuses to answer, it is asked the same question up to 10 times, and the most recent answer is used for scoring.
Still, an IQ-style benchmark captures only one slice of capability. It does not measure everything that matters in real-world AI use, such as coding ability, factual reliability, tool use, or performance in professional domains.
Learn More on the Voronoi App 
If you enjoyed today’s post, check out Global AI Adoption on Voronoi.
AI
Ranked: AI Models U.S. Businesses Pay For
OpenAI has long been the leader for paid usage by U.S. businesses, but Anthropic has closed the gap with tools like Claude Code and Cowork.
Published
2 days ago
on
April 23, 2026
Ranked: AI Models U.S. Businesses Pay For
See visuals like this from many other data creators on our Voronoi app. Download it for free on iOS or Android and discover incredible data-driven charts from a variety of trusted sources.
Key Takeaways
- OpenAI leads paid AI adoption among U.S. businesses at 35%, but Anthropic has surged to 30% in just over a year.
- Anthropic’s growth has been driven by enterprise tools like Claude Code and Cowork.
- Google, xAI, and others remain far behind, each used by less than 5% of businesses.
Anthropic is rapidly closing the gap with OpenAI in the race for paid AI adoption among U.S. businesses.
As of March 2026, 35% of companies pay for OpenAI’s models, compared to 30% for Anthropic—a sharp shift from early 2025, when the gap was nearly three times wider. The change highlights how quickly enterprise demand is consolidating around a small number of AI providers.
This chart, a part of Visual Capitalist’s AI Week sponsored by Terzo, uses anonymized spend data from over 50,000 U.S. businesses on the Ramp platform, capturing only paid subscriptions and excluding free-tier usage.
OpenAI Leads, But Anthropic Is Closing In Fast
OpenAI remains the most widely paid-for AI provider among U.S. businesses, reaching 35.2% of companies in March 2026. Anthropic sits just behind at 30.6%—a gap of only 4.5 percentage points.
The data table below shows the share of U.S. businesses paying for AI models from different providers from January 2023 to March of 2026:
| Share of U.S. Businesses Paying for an AI Subscription | ||||
|---|---|---|---|---|
| Date | OpenAI | Anthropic | xAI | |
| 1/1/2023 | 0.4% | 0.0% | 1.7% | 0.0% |
| 2/1/2023 | 1.5% | 0.0% | 1.6% | 0.0% |
| 3/1/2023 | 3.6% | 0.0% | 1.7% | 0.0% |
| 4/1/2023 | 5.7% | 0.0% | 1.8% | 0.0% |
| 5/1/2023 | 6.1% | 0.0% | 1.8% | 0.0% |
| 6/1/2023 | 5.9% | 0.0% | 1.9% | 0.0% |
| 7/1/2023 | 6.8% | 0.1% | 1.7% | 0.0% |
| 8/1/2023 | 7.2% | 0.1% | 1.7% | 0.0% |
| 9/1/2023 | 7.8% | 0.2% | 1.8% | 0.0% |
| 10/1/2023 | 8.1% | 0.3% | 1.8% | 0.0% |
| 11/1/2023 | 8.2% | 0.2% | 2.4% | 0.0% |
| 12/1/2023 | 9.3% | 0.3% | 2.4% | 0.0% |
| 1/1/2024 | 10.2% | 0.4% | 2.5% | 0.0% |
| 2/1/2024 | 10.2% | 0.4% | 2.6% | 0.0% |
| 3/1/2024 | 11.0% | 1.2% | 3.0% | 0.0% |
| 4/1/2024 | 10.6% | 1.4% | 3.3% | 0.0% |
| 5/1/2024 | 11.3% | 1.4% | 3.4% | 0.0% |
| 6/1/2024 | 11.0% | 1.5% | 3.2% | 0.0% |
| 7/1/2024 | 11.8% | 2.3% | 3.4% | 0.0% |
| 8/1/2024 | 12.5% | 2.5% | 3.5% | 0.0% |
| 9/1/2024 | 12.7% | 2.7% | 3.6% | 0.0% |
| 10/1/2024 | 13.7% | 3.0% | 3.7% | 0.0% |
| 11/1/2024 | 13.4% | 3.2% | 3.9% | 0.0% |
| 12/1/2024 | 14.8% | 3.6% | 4.0% | 0.0% |
| 1/1/2025 | 16.8% | 4.1% | 4.2% | 0.0% |
| 2/1/2025 | 18.2% | 4.4% | 4.2% | 0.2% |
| 3/1/2025 | 26.4% | 7.0% | 2.5% | 0.4% |
| 4/1/2025 | 32.0% | 7.9% | 3.2% | 0.5% |
| 5/1/2025 | 33.6% | 8.9% | 4.3% | 0.5% |
| 6/1/2025 | 33.4% | 9.6% | 4.0% | 0.6% |
| 7/1/2025 | 35.0% | 11.1% | 3.4% | 1.5% |
| 8/1/2025 | 36.5% | 12.1% | 3.0% | 1.5% |
| 9/1/2025 | 35.5% | 12.2% | 3.3% | 1.3% |
| 10/1/2025 | 35.8% | 14.3% | 3.3% | 1.6% |
| 11/1/2025 | 34.8% | 15.1% | 4.0% | 1.8% |
| 12/1/2025 | 36.8% | 16.7% | 4.3% | 1.9% |
| 1/1/2026 | 35.9% | 19.5% | 4.5% | 2.0% |
| 2/1/2026 | 34.4% | 24.4% | 4.7% | 1.9% |
| 3/1/2026 | 35.2% | 30.6% | 4.3% | 1.9% |
That gap looked very different a year ago. In January 2025, OpenAI was used by 16.8% of U.S. businesses while Anthropic sat at 4.1%, a spread of nearly 13 points. Anthropic has since grown more than sevenfold in 14 months, while OpenAI roughly doubled over the same period.
The remaining providers remain distant in paid business adoption. Google’s AI products—spanning Gemini, Vertex AI, and Workspace add-ons—have hovered between 3% and 4.5% of U.S. businesses for most of the past three years, barely moving despite heavy investment.
xAI has climbed from effectively zero in early 2024 to 1.9% in March 2026, a meaningful but still small footprint.
Claude Code and Cowork Drove the Anthropic Surge
Anthropic’s rapid rise in business adoption tracks its push into enterprise developer and knowledge-work tools.
Claude Code, the company’s coding assistant, and Cowork, its workflow collaboration platform, were both scaled aggressively across late 2025 and 2026—the period that coincides with the steepest part of Anthropic’s curve.
The pattern suggests that enterprise-native tooling, rather than general chatbot access, is now the key driver of paid seat growth. OpenAI has responded with its own developer coding tool, Codex, but Anthropic’s focus on developer workflows has clearly found traction in corporate procurement.
While Codex launched months after Claude Code, it has rapidly gained adoption among developers and knowledge workers, reaching four million active users as of April 21, 2026.