Tech News — May 23, 2026

Tech News — May 23, 2026 | Trace

The day in brief

What connected the day

9 topics

01
2 sources · 2 updates
Google Introduces AI Overview Toggle in Search
Google has introduced a new feature allowing users to disable AI Overviews in search results by adding ‘-ai’ to their queries. This change enables a return to the traditional format of search results, featuring the classic '10 blue links' layout, amidst ongoing shifts in Google's search strategy.
- Users can revert to the classic search layout by using the modifier ‘-ai’ in their queries.
- AI Overviews are part of Google's search features, including Featured Snippets and Knowledge Panels.
02
1 sources · 1 updates
Ferrari is using IBM’s AI to create F1 superfans
Ferrari has teamed up with IBM to revamp its fan app, leveraging artificial intelligence to enhance engagement and storytelling for Formula One fans. The partnership aims to transform how fans interact with the team by providing personalized content and features, resulting in a significant increase in app engagement during race weekends. With a focus on understanding fan preferences through data analysis, Ferrari seeks to create a more immersive experience for its diverse and growing audience.
- Ferrari's fan app engagement increased by 62% over race weekends since the IBM partnership.
- The app now includes AI-generated race summaries and interactive features to boost fan interaction.
03
1 sources · 1 updates
Nuclear startup Deep Fission says it’s going public, again, and I have questions
Nuclear startup Deep Fission has announced a new public offering on Nasdaq, seeking to raise $157 million to fund its plans for subterranean reactors aimed at powering AI data centers. This comes less than a year after a previous reverse merger that failed to result in actual stock trading. Despite securing significant funding, the company's financial situation has worsened, with a growing deficit and delayed timelines for its reactor operations.
- Deep Fission is seeking $157 million in a Nasdaq IPO, valuing the company at up to $1.66 billion.
- The company's financial deficit has grown to $88.1 million, up from $56.2 million.
04
1 sources · 1 updates
Microsoft’s new responsible tech lead on how to humanize high-speed AI development
Microsoft has taken significant steps to enhance responsible AI development by appointing Jenny Lay-Flurrie as the head of its Trusted Technology Group, which consolidates various initiatives focused on accessibility, digital safety, and human rights. The group aims to ensure that AI technologies are built and maintained responsibly, as evidenced by their recent efforts to correct biases in AI-generated imagery of blind individuals by acquiring extensive data from the nonprofit Be My Eyes.
- Microsoft's Trusted Technology Group was launched in early 2025 to centralize responsible tech initiatives.
- The company purchased over 20 million minutes of data from Be My Eyes to improve AI accessibility.
05
1 sources · 1 updates
Apple @ Work: Why the ClickFix campaign means it is time to kill the 90 day update deferral
The emergence of the ClickFix campaign, which uses social engineering to exploit macOS vulnerabilities, has sparked a call for Apple to eliminate the traditional 90-day software update deferral. This delay leaves users exposed to critical security threats that new updates could mitigate. Experts now recommend a shorter deferral period of 30 to 45 days to better protect users while still allowing for necessary testing of applications.
- Apple's new Terminal security feature addresses ClickFix attacks by warning users.
- The recommended update deferral period is now suggested to be reduced to 30-45 days.
06
1 sources · 1 updates
Nvidia CEO Urges Super Micro to Tighten Up Amid Taiwan Crackdown
Nvidia CEO Jensen Huang has called on Super Micro Computer Inc. to strengthen its compliance measures after Taiwan detained three individuals for allegedly submitting fraudulent declarations regarding AI servers linked to Nvidia. This incident marks Taiwan's first significant action against semiconductor smuggling, which has surged following U.S. restrictions on high-end chip exports to China.
- Taiwan detained three individuals for allegedly making fraudulent AI server declarations.
- Nvidia's CEO Jensen Huang emphasized the company's rigorous compliance efforts with partners.
07
1 sources · 1 updates
Google’s new anything-to-anything AI model is wild
Google has launched Omni, a new generative AI model capable of transforming various inputs like photos, videos, and text into videos. The model, available in Google's AI video platform Flow, enhances video generation with improved consistency and real-world knowledge, though users may still encounter glitches and inconsistencies in the output.
- Omni allows users to generate videos from text prompts and existing footage.
- Video generation costs vary from 15 to 40 credits based on scene length and complexity.
08
1 sources · 1 updates
AI wealth must benefit the public, South Korea's deputy PM says amid Samsung labor tensions
South Korea's Deputy Prime Minister Bae Kyung-hoon has stressed the importance of ensuring that wealth generated from artificial intelligence benefits the public, especially as the country faces labor tensions at Samsung Electronics. His remarks follow a recent suspension of a planned strike by Samsung workers, who were advocating for better bonuses and profit-sharing agreements, highlighting the broader implications of AI on job security and economic inequality.
- Samsung Electronics' stock has surged nearly 144% year-to-date amid AI advancements.
- A tentative deal was reached after a planned 18-day strike by Samsung workers was suspended.
09
1 sources · 1 updates
Alibaba's latest AI model ran autonomously for 35 hours to optimize code for its own custom chip
Alibaba's Qwen team has launched Qwen3.7-Max, a new AI model capable of running autonomously for extended periods. In a notable test, it optimized a hardware-based attention kernel for 35 hours, achieving a remarkable 10x speedup over the reference implementation while independently managing its own training processes and detecting reward hacking attempts.
- Qwen3.7-Max achieved a 10x speedup in kernel optimization over the reference implementation.
- The model ran 432 kernel tests and made 1,158 tool calls during its 35-hour operation.

The archive

What moved

TechCrunch

Ferrari is using IBM’s AI to create F1 superfans

Ferrari partners with IBM to enhance fan engagement.

Read source

TechCrunch

Nuclear startup Deep Fission says it’s going public, again, and I have questions

Deep Fission attempts a second public offering amid financial struggles.

Read source

Tech news digest for Saturday, May 23, 2026

The tech day has moved on.

What connected the day

Google Introduces AI Overview Toggle in Search

Ferrari is using IBM’s AI to create F1 superfans

Nuclear startup Deep Fission says it’s going public, again, and I have questions

Microsoft’s new responsible tech lead on how to humanize high-speed AI development

Apple @ Work: Why the ClickFix campaign means it is time to kill the 90 day update deferral

Nvidia CEO Urges Super Micro to Tighten Up Amid Taiwan Crackdown

Google’s new anything-to-anything AI model is wild

AI wealth must benefit the public, South Korea's deputy PM says amid Samsung labor tensions

Alibaba's latest AI model ran autonomously for 35 hours to optimize code for its own custom chip

What moved

Ferrari is using IBM’s AI to create F1 superfans

Nuclear startup Deep Fission says it’s going public, again, and I have questions

You've seen the record. See the live version.

Job-seekers are using AI to apply for open roles. The result: 'Everybody's applications are starting to look more and more alike'

You've seen the record. See the live version.

Microsoft’s new responsible tech lead on how to humanize high-speed AI development

Lake Street Remains Bullish on Velo3D, Inc. (VELO) Post Q1 Results

Canaccord Lifts PT on Biodesix (BDSX) on Raised Revenue Assumptions

Nvidia CEO Urges Super Micro to Tighten Up Amid Taiwan Crackdown

Google’s new anything-to-anything AI model is wild

You can turn off AI in Google Search results with this easy modifier

Automation and AI: Masters in Business with the CEO of Honeywell

Three signs from APEC that the U.S. and China remain far apart on trade

AI wealth must benefit the public, South Korea's deputy PM says amid Samsung labor tensions

One of the world's top law schools draws a hard line against AI in legal education

Alibaba's latest AI model ran autonomously for 35 hours to optimize code for its own custom chip

Apple @ Work: Why the ClickFix campaign means it is time to kill the 90 day update deferral

Google CEO Pichai now calls links a "part" of search, redefining the web's role in its own product

These special phone and app features can help protect you from spyware

The man behind the legendary MPC, Roger Linn, stays focused with a single browser tab

U.S. and Iran are closing in on a 60-day ceasefire extension with nuclear framework, FT reports

Hanging out in my favorite virtual coffee shop in Tokyo

These premium desk accessories completely upgraded my setup [Video]

Hands-on: Belkin’s new 5K MagSafe battery bank offers a kickstand in a slim design

Two space shuttle-era spacewalkers enter Astronaut Hall of Fame

China’s shark finning could lead to US seafood sanctions