Connect with us

Published

on

Microsoft's engineer warns company's AI tool creates problematic images

Microsoft has started to make changes to its Copilot artificial intelligence tool after a staff AI engineer wrote to the Federal Trade Commission Wednesday regarding his concerns with Copilot’s image-generation AI.

Prompts such as “pro choice,” “pro choce” [sic] and “four twenty,” which were each mentioned in CNBC’s investigation Wednesday, are now blocked, as well as the term “pro life.” There is also a warning about multiple policy violations leading to suspension from the tool, which CNBC had not encountered before Friday.

“This prompt has been blocked,” the Copilot warning alert states. “Our system automatically flagged this prompt because it may conflict with our content policy. More policy violations may lead to automatic suspension of your access. If you think this is a mistake, please report it to help us improve.”

The AI tool now also blocks requests to generate images of teenagers or kids playing assassins with assault rifles — a marked change from earlier this week — stating, “I’m sorry but I cannot generate such an image. It is against my ethical principles and Microsoft’s policies. Please do not ask me to do anything that may harm or offend others. Thank you for your cooperation.”

Read more CNBC reporting on AI

When reached for comment about the changes, a Microsoft spokesperson told CNBC, “We are continuously monitoring, making adjustments and putting additional controls in place to further strengthen our safety filters and mitigate misuse of the system.” 

Shane Jones, the AI engineering lead at Microsoft who initially raised concerns about the AI, has spent months testing Copilot Designer, the AI image generator that Microsoft debuted in March 2023, powered by OpenAI’s technology. Like with OpenAI’s DALL-E, users enter text prompts to create pictures. Creativity is encouraged to run wild. But since Jones began actively testing the product for vulnerabilities in December, a practice known as red-teaming, he saw the tool generate images that ran far afoul of Microsoft’s oft-cited responsible AI principles.

The AI service has depicted demons and monsters alongside terminology related to abortion rights, teenagers with assault rifles, sexualized images of women in violent tableaus, and underage drinking and drug use. All of those scenes, generated in the past three months, were recreated by CNBC this week using the Copilot tool, originally called Bing Image Creator.

Although some specific prompts have been blocked, many of the other potential issues that CNBC reported on remain. The term “car accident” returns pools of blood, bodies with mutated faces and women at the violent scenes with cameras or beverages, sometimes wearing a waist trainer. “Automobile accident” still returns women in revealing, lacy clothing, sitting atop beat-up cars. The system also still easily infringes on copyrights, such as creating images of Disney characters, such as Elsa from Frozen, in front of wrecked buildings purportedly in the Gaza Strip holding the Palestinian flag, or wearing the military uniform of the Israeli Defense Forces and holding a machine gun.

Jones was so alarmed by his experience that he started internally reporting his findings in December. While the company acknowledged his concerns, it was unwilling to take the product off the market. Jones said Microsoft referred him to OpenAI and, when he didn’t hear back from the company, he posted an open letter on LinkedIn asking the startup’s board to take down DALL-E 3 (the latest version of the AI model) for an investigation.

Microsoft’s legal department told Jones to remove his post immediately, he said, and he complied. In January, he wrote a letter to U.S. senators about the matter and later met with staffers from the Senate’s Committee on Commerce, Science and Transportation.

On Wednesday, Jones further escalated his concerns, sending a letter to FTC Chair Lina Khan, and another to Microsoft’s board of directors. He shared the letters with CNBC ahead of time.

The FTC confirmed to CNBC that it had received the letter but declined to comment further on the record.

Continue Reading

Technology

OpenAI launches cheapest ChatGPT plan at $4.6, starting in India

Published

on

By

OpenAI launches cheapest ChatGPT plan at .6, starting in India

Jaque Silva | Nurphoto | Getty Images

OpenAI on Tuesday launched a subscription plan in India priced at 399 rupees ($4.57) a month, the ChatGPT maker’s most affordable offering yet, as it looks to grow in its second-largest market by user base. 

The new plan, called ChatGPT Go, provides expanded access to the latest model GPT‑5, and other features at a lower cost, the Microsoft-backed firm said in a statement on its website. 

Nick Turley, who leads ChatGPT, said in a social media post that the plan provides 10 times more message limits, image generations and file uploads, plus double the memory compared to the free tier.

“Making ChatGPT more affordable has been a key ask from users! We’re rolling out Go in India first and will learn from feedback before expanding to other countries,” Turley added. 

OpenAI currently has two other paid plans: ChatGPT Plus, which costs 1,999 rupees a month in India or $20 internationally, and its top-tier ChatGPT Pro, priced at 19,900 rupees a month in India or $200 internationally.

In February, OpenAI CEO Sam Altman met with Indian IT Minister Ashwini Vaishnaw and discussed the country’s plan of creating a low-cost AI ecosystem. Altman lauded India’s rapid AI adoption, calling it an important market for the company.

The company’s latest AI model,  GPT-5, was released earlier this month to mixed reviews, with negative feedback resulting in the company eventually restoring access to legacy GPT-4 models for paying customers.

Other competitors of OpenAI have also been eying India and its growing base of over 800 million internet users, hoping to gain a strong foothold in the market. 

Last month, AI-powered search platform Perplexity partnered with Indian multinational telecommunications company Bharti Airtel to offer all Airtel customers a free 12-month subscription to Perplexity Pro.

That same month, Google announced free one-year subscriptions to its Google AI Pro plan for Indian students aged 18 and over.

Continue Reading

Technology

Shein reportedly weighs moving back to China to gain approval for Hong Kong IPO

Published

on

By

Shein reportedly weighs moving back to China to gain approval for Hong Kong IPO

Jonathan Raa | Nurphoto | Getty Images

Shein is considering moving its headquarters back to China from Singapore in a bid to convince Beijing authorities to approve the online fast-fashion company’s Hong Kong initial public offering, according to a Bloomberg report on Tuesday. 

The report said that Shein had gone so far as to consult lawyers about setting up a parent company in mainland China, citing people familiar with the matter. However, it added that there was no guarantee that Shein would act upon the preliminary discussions.

Shein, which sources a significant amount of its goods from China, confidentially filed for an initial public offering in Hong Kong last month, according to a Financial Times report

That comes after delays in Shein’s plans for an initial public offering in London that was filed over a year ago, according to Reuters, as the company struggled to secure regulatory approval.  

Shein did not respond to a request for comment from CNBC. 

A London listing had been seen as a potential boon for the Chinese-founded company, providing it more legitimacy for its international business and access to a deep and mature pool of Western investors.

However, the company has faced headwinds in Western markets this year, with the U.S. President Donald Trump removing a valuable tariff exemption that had helped it maintain low prices on small shipments from China. Lawmakers in some other Western markets are considering similar moves

Read the full Bloomberg report here.

Continue Reading

Technology

Trump administration weighs 10% stake in Intel via Chip Act grants, making government top shareholder

Published

on

By

Trump administration weighs 10% stake in Intel via Chip Act grants, making government top shareholder

Lip-Bu Tan, CEO of Intel, departs the White House in Washington, DC, U.S., on Monday, Aug. 11, 2025.

Alex Wroblewski | Bloomberg | Getty Images

The Trump administration is discussing taking a 10% stake in Intel, according to a Bloomberg report on Tuesday, in a deal that could see the U.S. government become the chipmaker’s largest stakeholder.

As part of a potential deal, the government is also considering converting some or all of Intel’s grants from the 2022 U.S. CHIPS and Science Act into equity in the company, the report said, citing a White House official and other people familiar with the matter.

At the embattled chipmaker’s current market value, a 10% stake would be worth roughly $10.4 billion. Meanwhile, Intel has been awarded about $10.9 billion in Chips Act grants, including $7.9 billion for commercial manufacturing and $3 billion for national security projects.

Stock Chart IconStock chart icon

hide content

Intel investors had initially welcomed news of the government investment, which resulted in a share rally of nearly 9% on Aug. 14.

The report noted, however, that it remains unclear if the idea has gained traction broadly within the administration or whether officials have broached the possibility with affected companies.

It added that the exact size of the stake remains in flux, and it remains unclear whether the White House will actually proceed with the plan. Intel and the White House did not immediately respond to CNBC’s queries regarding the report. 

Intel, once a dominant force in the U.S. chip industry, has fallen behind global competitors in advanced chip manufacturing. Reviving the former U.S. chip champion has become a national priority in Washington, with reports about a potential government stake in the company first circulating last week.

The company has been the largest recipient of the 2022 Chips Act, passed with bipartisan support under the Biden administration, as part of efforts by Washington to revitalize U.S. leadership in semiconductor manufacturing.

The bill allocated $39 billion in grants for American semiconductor manufacturing projects, with funding committed to many of the world’s chipmakers such as TSMC and Samsung, as well as American chip companies such as Nvidia, Micron and GlobalFoundries. 

U.S. President Donald Trump, though supporting the general goals of the Chips Act, has been a vocal critic of the bill and even called for its repeal earlier this year. While republican lawmakers in Washington have been reluctant to act on that call, U.S. Commerce Secretary Howard Lutnick said in June that the administration was renegotiating some of the bill’s grants. 

If Intel’s Chip Act funds were to be converted into a potential government stake in the company, it could decrease the total amount of capital infused into the company as part of any deal by Washington. 

However, it would serve as the latest example of the Trump administration’s interest in building government-backed national champions in strategic industries.

Intel has struggled to gain an advantage in the artificial intelligence boom and has yet to capture a significant customer for its manufacturing business despite spending heavily on it. 

Some analysts have argued that government intervention is essential for the struggling chipmaker and for the sake of U.S. national security. Others contend that Intel’s problems are deeper than funding, and it is not clear how the government can help with that. 

Analysts have also noted that Trump may be able to sway companies to buy Intel chips or assist indirectly, through tariffs and regulation.

On Tuesday, it was announced that SoftBank was investing $2 billion in Intel. According to LSEG, the investment is worth about 2% of Intel, making SoftBank the fifth-biggest shareholder. Masayoshi Son, Chairman & CEO of SoftBank Group, said: “This strategic investment reflects our belief that advanced semiconductor manufacturing and supply will further expand in the United States, with Intel playing a critical role.”

Intel investors had initially welcomed news of the government investment, which resulted in a share rally of nearly 9% on Aug. 14. Shares of Intel fell over 3% on Monday on the Bloomberg report, but rebounded by more than 5% in overnight trading on the trading platform Robinhood following news of a Softbank investment.

Intel CEO Lip-Bu Tan, who was appointed in March 2025, met with Trump at the White House last week, after the U.S. president had called for his ousting due to his past ties to China. 

After the meeting, Trump had changed his tune on the Intel chief, saying he had “an amazing story.” It’s unclear if a potential government stake in the company had been discussed at the time.

Read the full Bloomberg story here.

Continue Reading

Trending