Connect with us

Technology

Microsoft and OpenAI sued the New York Times for ‘billions’

Published

on

Microsoft and OpenAI sued the New York Times for 'billions'
Get this image on: Flickr | Creator: Wally Gobetz

Report By: Tom GerkenĀ for the BBC

In a legal skirmish that unfolded on Wednesday, the New York Times, a prominent US news entity, initiated legal action against OpenAI, the proprietor of ChatGPT. This legal tussle also entangles Microsoft, who stands accused alongside OpenAI, with the lawsuit contending that these entities bear responsibility for incurring “billions of dollars” in damages.

At the crux of the matter lies the training of ChatGPT, a sophisticated language model, allegedly involving the unauthorized utilization of copyrighted material from the New York Times. ChatGPT, among other expansive language models, acquires knowledge through meticulous analysis of copious datasets, often culled from online sources.

OpenAI and Microsoft have been tight-lipped about the lawsuit when approached by the BBC for commentary on the matter.

The crux of the legal complaint posits that ChatGPT’s augmentation entailed the unauthorized incorporation of “millions” of articles from the New York Times, thereby enhancing its cognitive capabilities. Furthermore, the lawsuit suggests that ChatGPT, having absorbed this substantial content, is now positioned as a direct competitor to the New York Times as a reliable purveyor of information.

A compelling allegation within the legal dossier is that when prompted about current affairs, ChatGPT occasionally regurgitates “verbatim excerpts” sourced from New York Times articles, content inaccessible without a subscription fee. This, according to the lawsuit, permits readers to access New York Times material without remitting the requisite subscription charges. Consequently, the newspaper contends that it is not only losing revenue from subscriptions but also from advertising clicks, as users can bypass the paywall.

The lawsuit introduces a pertinent example involving the Bing search engine, which incorporates ChatGPT’s capabilities. The claim is that Bing’s results, powered by ChatGPT, derive from a New York Times-owned website, yet fail to provide direct links or referral mechanisms that contribute to income generation. Noteworthy is Microsoft’s substantial investment, exceeding $10 billion (Ā£7.8 billion), in OpenAI.

This legal clash, lodged in a Manhattan federal court, lays bare the New York Times’ fruitless attempt in April to engage Microsoft and OpenAI in seeking “an amicable resolution” concerning the alleged copyright infringement.

Multiple lawsuits

In the aftermath of a period at OpenAI, one month witnessed a whirlwind of chaos, marked by the abrupt dismissal and subsequent reinstatement of co-founder and CEO Sam Altman over the span of a few days.

The seismic event of Altman’s removal sent shockwaves through industry insiders, leaving them astounded. The repercussions reached a crescendo when OpenAI’s staff, in response to Altman’s ousting, issued dire threats of mass resignations unless he was promptly reinstated.

However, beyond the internal tumult, the organization now grapples with a cluster of lawsuits, all initiated in the year 2023.

In September, a legal echo resonated as a consortium of U.S. authors, featuring luminaries like George RR Martin of Game of Thrones fame and John Grisham, brought forth a copyright infringement case mirroring previous litigation. This legal volley followed comedian Sarah Silverman’s legal action in July. Concurrently, authors Margaret Atwood and Philip Pullman endorsed an open letter during the same month, urging AI companies to provide just compensation for the utilization of their creative works.

Adding to the legal maelstrom, OpenAI is entangled in yet another lawsuit alongside Microsoft and the programming platform GitHub. A group of computing experts contends that their code was surreptitiously employed without authorization in training an AI entity named Copilot.

Parallel to these legal maneuvers, developers of generative AI, a strain of artificial intelligence capable of crafting media based on textual prompts, find themselves entwined in legal entanglements. In January, artists initiated legal actions against Stability AI and Midjourney, asserting that these text-to-image generators rely exclusively on training from copyrighted artwork.

As it stands, the legal horizon remains unresolved, with none of these lawsuits reaching a conclusive resolution. OpenAI navigates a complex landscape, grappling not only with internal upheaval but also with a multifaceted array of legal challenges.

This story was first published by theĀ BBC

More: Sam Altman will return as the CEO of OpenAI

Technology

Investors are being “AI washing” everywhere

Published

on

Investors are being AI washing everywhere

Remaining competitive in the finance sector is crucial, as numerous companies are swiftly adopting artificial intelligence (AI) to cut costs and streamline operations. However, two companies recently faced trouble when the United States Securities and Exchange Commission (SEC) accused them of overstating their use of AI, marking the worldā€™s initial significant effort in combating what’s known as ā€œAI washingā€. Delphia (USA) Inc and Global Predictions Inc had boasted about utilizing AI to develop investment strategies, but the SEC discovered their assertions lacked evidence. There’s considerable speculation surrounding AI, particularly with the emergence of generative technology app ChatGPT. Yet, amidst the buzz, AI washing is becoming increasingly prevalent. In addition to overstating or misrepresenting their AI capabilities, companies may exaggerate the capabilities of AI algorithms or create the impression that AI plays a larger role than it actually does.

Whatā€™s so good about AI?

Integrating AI into business operations offers numerous advantages. It can streamline processes, swiftly analyze complex data to expedite decision-making, and assist organizations in staying ahead in a swiftly evolving and competitive market. Promoting the utilization of AI can present a company as cutting-edge and technologically advanced, even if the reality doesnā€™t align. The concept of AI washing isnā€™t entirely novel; it parallels the notion of greenwashing, where companies feign eco-friendliness to attract investors and consumers. It entails embellishing ordinary technology with sophisticated AI jargon such as ā€œmachine learning,ā€ ā€œneural networks,ā€ ā€œdeep learning,ā€ and ā€œnatural language processingā€ to appear more innovative than they truly are.

AI and the finance sector

AI washing thrives in finance and investment due to the industryā€™s high stakes, fierce competition, and the alluring allure of technology-driven solutions. AI’s algorithms can analyze vast datasets, improve predictability, and unveil hidden patterns in financial data. Moreover, AI’s real-time processing capabilities enable agile adaptation to market fluctuations. The intricacy of financial products allows firms to mask the truth behind flashy AI assertions, while the absence of regulation exacerbates the issue. Despite AI’s remarkable capabilities, it is not immune to drawbacks, including ethical concerns, susceptibility to cyber attacks and manipulation, and the lack of transparency in decision-making processes. Supporters of AI-related investments range from inexperienced retail investors to seasoned institutional players. Such interest has prompted venture capital firms to allocate more capital to AI startups last year than they had previously.

Lack of regulation

However, in the absence of clear guidelines, companies can exploit loopholes and deceive investors. This lack of oversight undermines trust and credibility within the industry. Moreover, AI washing may hinder innovation. If investors harbor doubts about AI, they’re less inclined to invest in genuine AI-powered solutions, potentially impeding the advancement of truly revolutionary technologies.

Dealing with AI washing is crucial, echoing the cautionary narrative of the dot-com bubble. Similar to the exaggerated assurances and speculative enthusiasm surrounding internet companies that precipitated market turmoil and investor skepticism in the late 1990s, the hype surrounding AI capabilities in finance poses analogous risks.

AI washing could prompt investors to inject funds into AI-related ventures without fully grasping the risks or potential limitations, potentially exposing them to financial losses when the bubble bursts. While the European Union AI Act stands as the world’s inaugural regulation governing the use, development, disclosure, and oversight of AI, Australia lacks specific laws in this regard. Presently, regulation falls under the purview of the Corporations Act. ASIC is currently exploring methods to regulate AI, including devising penalties for AI washing. Holding companies accountable for providing accurate information regarding technology applications helps preserve the integrity of financial markets and ensures equity for investors.

How to spot AI washing

Here are some tips for investors and consumers to avoid falling victim to AI washing:

1. Verify registration status and credentials

Before purchasing or investing in any product or service claiming AI capabilities, verify the registration status and credentials of the investment company by checking the professional register. Ensure they have no disciplinary history by consulting the Australian Securities and Investment Commission register.

2. Be cautious with AI-focused investments

Investing in AI-driven companies may appear promising, but exercise caution with firms that boast about their ā€œrevolutionaryā€ or ā€œindustry-leadingā€ AI without offering specifics. What exactly makes their AI revolutionary? What problems does it address? Companies that depend on vague buzzwords without providing concrete details likely exaggerate their capabilities.

3. Boost your knowledge

Gain a basic understanding of AI and machine learning. Familiarize yourself with common AI techniques and terms commonly used in finance. There are numerous free online resources available for beginners.

4. Ask questions

Do not rely solely on AI-generated information for making investment decisions. AI-generated data might be inaccurate or biased. Ask financial advisors and companies about their particular AI implementation. What type of data do they utilize? How are their algorithms trained? What limitations does their technology have?

5. Be sceptical of high returns with little to no risk

Be cautious of financial products that promise high returns with low risk, particularly those boasting AI-powered success. This is a common warning sign of AI washing. Don’t solely trust a company’s assertions ā€“ conduct independent research by monitoring financial news or reviewing companies’ regulatory filings before investing.

More: 63 percent of Indian businesses believe that Generative AI is crucial for their sustainability efforts, according to a study

Continue Reading

Technology

Xi Jinping tells Joe Biden not to curb Chinaā€™s technology sector

Published

on

Xi Jinping tells Joe Biden not to curb Chinaā€™s technology sector

 

The top leaders of the United States and China engaged in a two-hour phone call on Tuesday, establishing the context ahead of US Treasury Secretary Janet Yellen’s forthcoming meetings with Chinese counterparts in Guangzhou and Beijing from April 4 to April 9.

During the call, Chinese President Xi Jinping conveyed to US President Joe Biden that the US has implemented a series of measures aimed at curtailing China’s trade and technological advancement. Xi highlighted that the US continues to expand its sanctions against Chinese entities, asserting that such actions do not reduce risk but rather create additional risks.

Xi emphasized that if the US demonstrates willingness to pursue mutually beneficial cooperation and partake in China’s developmental benefits, China’s doors will remain open for engagement. However, he warned that if the US persists in its efforts to restrain China’s progress in high technology and deny China its rightful developmental opportunities, China will not passively observe.

Xi’s remarks were prompted by the US Commerce Department’s recent revision of chip export regulations on March 29, which were initially announced in October. These revisions are intended to impede China’s access to US-made artificial intelligence (AI) chips and chip manufacturing equipment.

The revised regulations, slated to come into effect on Thursday, state that restrictions on chip exports to China also extend to laptops containing advanced built-in AI chips.

The updated regulations mandate a license for the export of digital computers and electronic assemblies with adjusted-peak-performance (APP) exceeding 70 weighted teraflops (WT), as well as for software tailored for the development or production of digital computers with APP surpassing 24 WT. The US will implement a presumption-of-denial policy for license applications concerning exports to destinations such as Macau and D:5 countries like Russia and China. It will adopt a case-by-case review approach for AI chip exports to China.

Reports from some media outlets suggest that Nvidiaā€™s RTX 4090D graphics cards and H20 (Hopper) data-center accelerators, boasting computing power of 73.5 and 74 teraflops, respectively, will be impacted by the updated regulations. The demand for RTX 4090D graphics cards has surged in recent days as vendors in Shenzhen are hoarding them.

The Chinese Commerce Ministry has expressed strong opposition to the revised US chip export rules, criticizing the US for stretching the concept of national security, erecting more barriers to normal trade cooperation between Chinese and American firms, imposing heavier compliance burdens, and introducing significant uncertainties in the global chip sector.

Negative facts

Following the recent phone call between Xi and Biden, both Beijing and Washington characterized the dialogue as candid and constructive. They mutually agreed to maintain communication and instructed their respective teams to follow up on the issues discussed during the leaders’ meeting in San Francisco last November.

The Chinese side expressed its welcome for the forthcoming visits to China by Yellen and US Secretary of State Antony Blinken in the near future.

During the call, Biden reiterated ongoing concerns regarding China’s trade policies and non-market economic practices, which the US claims unfairly impact American workers and families. He affirmed that the US will persist in taking necessary measures to prevent advanced US technologies from being misused to jeopardize American national security, while ensuring that trade and investment are not unduly restricted.

Xi acknowledged that the China-US relationship is showing signs of stabilization, but also noted the emergence of certain negative factors. He emphasized the need for attention from both sides to address these concerns.

Wu Xinbo, the executive dean of the Institute of International Studies at Fudan University, remarked that Xi and Biden utilized the phone call to articulate their primary concerns regarding Sino-US relations and establish the trajectory for the next phase of development between the two nations.

According to Wu, China expects the US to maintain its credibility and adhere to the commitments mutually agreed upon in San Francisco through tangible actions. He noted that the Biden administration has been endeavoring to constrain China under the pretext of “competition” to satisfy its domestic political objectives. Wu highlighted Beijing’s dissatisfaction with the US’s continual advancement of its Indo-Pacific strategy and frequent provocations by supporting the Philippines in the South China Sea.

Diao Daming, a professor at the School of International Studies at Renmin University of China, emphasized that China desires the US to fulfill the promises made in San Francisco last November, which have yet to materialize. He stressed the importance of the US recognizing and managing the risks inherent in US-China relations, citing conflicts arising from US sanctions against Chinese firms and its restrictions on chip exports to China.

Support for Russia

During the phone conversation with Xi, Biden expressed concerns regarding China’s backing of Russia’s defense industrial base and its potential ramifications on European and transatlantic security.

In a meeting with Chinese Foreign Minister Wang Yi held in Bangkok on January 26-27, US White House national security advisor Jake Sullivan stated that China had supplied heavy trucks and other equipment to Russia’s defense industrial base. Sullivan highlighted that the US had observed assistance from Chinese companies aiding Russia in rebuilding its defense industry, a development he deemed as potentially signaling imperialistic intentions in Europe. Sullivan emphasized that this issue touches upon a core national security interest of the United States.

Subsequently, on February 23, the Biden administration-imposed sanctions on 93 entities from Russia, China, Turkey, and other countries for their support of Russia’s military efforts in Ukraine.

In February, US officials disclosed to the media that the White House is contemplating imposing limitations on the importation of Chinese electric vehicles, citing escalating US apprehensions regarding data security. In December of the previous year, Biden had indicated the possibility of increasing tariffs on Chinese electric vehicles.

On March 27, Yellen remarked that the surplus capacity buildup in China’s burgeoning industries such as solar energy, electric vehicles, and lithium-ion batteries could adversely affect other nations.

Yellen is scheduled to hold meetings with Chinese Vice Premier He Lifeng in Guangzhou on April 5-6 and with Premier Li Qiang in Beijing on April 7.

More: Rakuten Mobile in Japan is making progress with Open RAN technology

Continue Reading

Technology

Rakuten Mobile in Japan is making progress with Open RAN technology

Published

on

Rakuten Mobile in Japan is making progress with Open RAN technology

 

Rakuten Mobile, the emerging mobile network provider in Japan, anticipates reaching a turning point in its finances as subscriber growth moves closer to breaking even in terms of operating cash flow by 2024. Achieving this breakeven point will mark a significant milestone for the Open RAN (Radio Access Network) standard, which enables various companies to contribute different components to a telecommunications network. Open RAN offers telecom service providers increased flexibility and autonomy, reducing dependence on major suppliers of proprietary systems like Huawei in China, Ericsson in Sweden, and Nokia in Finland, as well as smaller domestic players such as Samsung in South Korea, and NEC and Fujitsu in Japan.

Mavenir, a US-based telecom software company, estimates that a cloud-based Open RAN network can reduce deployment and operational costs by 37% over five years compared to proprietary architectures. This assessment comes at a time when there are allegations that major telecom equipment manufacturers, particularly Ericsson, are trying to manipulate Open RAN, with it being touted as a “Huawei killer” in Washington, DC. However, for companies like Rakuten and other supporters of Open RAN, the focus is on proving the technology’s commercial viability. Rakuten Mobile, which is exclusively dedicated to Open RAN, had approximately 6.5 million subscribers by the end of December 2023. If the current growth rate seen in 2024 continues, the company is expected to reach its target of 8-10 million subscribers by the end of the year.

Rakuten Mobile’s subscriber count lags behind Japanese market leaders NTT Docomo, KDDI (au), and Softbank, which had 89.2 million, 66.9 million, and 40.1 million subscribers respectively by the end of 2023. However, it does highlight that the combination of the low-cost Open RAN business model and discounted pricing can offer customers better deals. This aligns with the original goal of then-prime minister Yoshihide Suga’s government in 2018, which aimed for increased competition and the reduction of market domination by the three major companies, with a focus on providing cheaper mobile phone services.

Founded in January 2018, Rakuten Mobile obtained a license to construct and operate a fourth-generation (4G) mobile telecommunications network a year later. Its 5G service was launched in September 2020. By February 2022, the network had expanded its coverage to reach 96% of Japan’s population. Presently, it covers almost 100% of the population and 99% of the nation’s territory.

Service plans offered by Rakuten Mobile vary, but numerous comparisons indicate that Rakuten’s pricing is significantly lower than that of Japan’s major carriers, especially after the latter reduced their fees in response to Rakuten’s entry. Rakuten’s rates are also competitive with or cheaper than those of discount providers such as UQ mobile, Y!mobile, J:COM, and BIGLOBE.

In terms of quality, Rakuten has generally kept pace with its competitors. According to Opensignal, a telecom technology and market research company, Rakuten Mobile ranks as the top carrier in Japan for 5G download and upload speeds. It also ranks second in terms of consistent quality and live video experience, trailing only behind Softbank and surpassing both NTT Docomo and KDDI.

Rakuten Mobile is the brainchild of Hiroshi Mikitani, the founder and CEO of the Japanese e-commerce giant Rakuten. Despite enduring significant losses, Mikitani has remained committed to the project, a level of dedication uncommon among most CEOs.

Over the past five years, Rakuten has reported cumulative net losses totaling 993.5 billion yen (US$6.6 billion) on a consolidated basis, with the trend likely to continue into 2024. This prolonged financial struggle has caused investors to lose confidence in the company, resulting in a steep decline of approximately 70% in its stock value from March 2021 to June 2023. However, there has been a partial recovery since then, reclaiming more than a third of the lost value, mainly due to declining operating losses in Rakuten Mobile.

Mobile telecommunications represents one of Rakuten’s three primary business divisions, alongside online shopping and finance.

Rakuten Mobile’s operating cash flow, which comprises operating profit plus depreciation and amortization, has shown improvement, decreasing from negative 70.6 billion yen in the fourth quarter of 2022 to negative 29.5 billion yen in the fourth quarter of 2023. Management anticipates that it will reach breakeven by the end of this year and turn positive in 2025.

With Rakuten Mobile’s domestic network largely completed, capital spending is declining, while quarterly data indicates a rapid increase in the number of subscribers throughout 2023. However, even if the company achieves management’s targets, it is expected to continue exerting a negative impact on Rakuten’s overall profitability for another two years.

In March 2020, Rakuten Mobile and NEC, a leading Japanese telecommunications equipment manufacturer, announced the commencement of production for jointly developed 5G radio units at NEC’s factory in Fukushima. These radio units adhere to open architecture standards, feature compact lightweight antennas, and are characterized by low power consumption, as highlighted by NEC at the time of the announcement.

Following the collaboration with NEC, Rakuten Mobile proceeded to construct 5G base stations for its fully virtualized cloud computing-based Open RAN mobile network. Mavenir, a US-based company claiming to be the sole end-to-end provider of cloud-native network software for telecom service providers, contributed its latest voice and messaging technology.

According to American software company Red Hat, virtualized radio access networks (vRANs) allow telecommunications operators to execute their baseband functions as software. One of the main advantages is that RAN functions no longer necessitate specialized proprietary hardware but can instead operate on standard servers.

ACG Research estimates that virtual RANs operated through cloud computing can decrease network operators’ total cost of ownership by 44%. Rakuten Mobile itself states that its network costs 40% less to construct and 30% less to operate compared to certain single-vendor proprietary mobile networks.

Rakuten Mobile’s Open RAN ecosystem involves contributions from various equipment suppliers and software developers. Airspan, based in the US, offers its vRAN (virtualized radio access networks) hardware and software platform to Rakuten Mobile, collaborating on joint marketing efforts for the Rakuten communications platform. Korea Microwave (KMW) and Microelectronics Technology Inc (MTI) from Taiwan provide radio units, while Qucell from South Korea and Sercomm from Taiwan supply small cell technology. Small cells are compact radio access points that enhance cellular network coverage, particularly in densely populated urban areas and high-traffic locations.

Prose Technologies, headquartered in Ireland, supplies and supports radio units, distributed antenna systems, and other components. Druid Software, also based in Ireland, has deployed its private cellular network software on Rakuten’s cloud platform, which is built on Intel technology. Supermicro, headquartered in San Jose, provides servers and collaborates with Rakuten on deploying automated cloud-native network solutions. Additionally, companies like Dell, Google, Nokia, and Tata Communications from India are also involved in collaborations with Rakuten Mobile.

In August 2021, Rakuten and German telecom service provider 1&1 joined forces to establish Europe’s inaugural virtualized Open RAN mobile network. By December 2023, 1&1 commenced operations with a core network provided by Mavenir, while Rakuten took charge of the end-to-end integration of the network, involving around 80 other vendors.

In areas where the 1&1 network remains incomplete, coverage is ensured through nationwide roaming on the Telefonica network. Additionally, starting this summer, roaming services will also be accessible on the Vodafone network.

The prospects for Open RAN have elicited considerable optimism, but there have also been significant doubts about its viability, as well as concerns about the sustainability of Rakuten Mobile. However, after years of dedication and substantial investment, Rakuten Mobile is gradually gaining market share in Japan and making strides in Europe. Should Rakuten or 1&1 face eventual failure, it is likely that larger telecom carriers would assume control of their operations.

More: 63 percent of Indian businesses believe that Generative AI is crucial for their sustainability efforts, according to a study

Continue Reading

Trending