Edition # 7

Hello Readers,

In the latest edition of DDP, a convergence of pivotal topics unfolds, shaping the trajectory of artificial intelligence (AI) and its global impact. From Congress's landmark decision on TikTok's fate to Saudi Arabia's bold investment in AI superpower status, and collaborative efforts by Chinese and US tech giants to establish global AI standards, and new measures clarifying security review requirements for cross-border data transfer in China. This newsletter edition delves into the interconnected narratives driving AI innovation and regulation. Amidst Big Tech's billion-dollar investments in AI infrastructure, the implications for society are profound and far-reaching.?

Happy reading!

Tech

Congress passes TikTok sell-or-ban bill, but legal battles loom

USA Today: ?In a significant move in the realm of social media regulation, the Senate has passed a bill aimed at restricting the popular video app TikTok. Embedded within a $95 billion foreign aid package, this legislation mandates ByteDance, TikTok's parent company based in Beijing, to sell the app within a year. Failure to comply would result in the app being barred from U.S. app stores and web hosting services. The bipartisan support for this measure was evident, with the Senate passing it 79-18 and the House approving the TikTok provision 360-58. President Joe Biden swiftly signed the bill into law.

Advocates of the policy argue that TikTok presents a national security risk, expressing concerns about potential Chinese government espionage and propaganda dissemination through the app. Despite TikTok's assertions that it has not been approached by the Chinese government for American users' data and would not comply if such a request were made, legislators remain wary. The company contends that the legislation infringes upon Americans' freedom of speech and could detrimentally impact small businesses reliant on TikTok for visibility.

Although there is no public evidence suggesting TikTok is actively used for spying on U.S. citizens, reports indicate challenges in fully separating TikTok's American operations from its Chinese parent company. Responding to the new law, a TikTok spokesperson labeled it "unconstitutional," with CEO Shou Zi Chew vowing to challenge it legally, citing confidence in prevailing based on facts and constitutional grounds.

What happens when we train our AI on social media? While cost effective and efficient, training AI models on social media carries some major risk.

FastCompany: Recent advancements in the fusion of social media and AI are grabbing attention, highlighted by Reddit teaming up with Google to let Google's AI learn from Reddit's content. Meanwhile, tech companies are revisiting old social media platforms like Photobucket. But amidst these developments, a crucial question arises: Is it wise to train AI using social media content?

While social media has been a treasure trove of data for AI training due to its sheer volume and accessibility, there are significant concerns. Issues such as user privacy and the prevalence of toxic content and misinformation on these platforms are causing apprehension. Critics worry that AI trained on such data could inadvertently perpetuate these negative traits.

Efforts are underway to tackle these challenges, with solutions in the works to counteract the spread of misinformation and toxicity in AI trained on social media data. For more details, check out the full article.

AI

To the Future: Saudi Arabia spends big to become an AI superpower:

The Business Times: Tech companies are increasingly cozying up to Saudi Arabia as the kingdom aims to assert itself as a major player in the field of artificial intelligence (AI), pouring staggering amounts of money into the endeavor. This year, Saudi Arabia established a hefty $100 billion fund earmarked for investments in AI and other tech ventures. This ambitious spending spree is part of "Vision 2030," a long-term initiative spearheaded by Crown Prince Mohammed bin Salman since 2016.

If successful, Crown Prince Mohammed's vision could position Saudi Arabia as a formidable contender in the global AI race, competing with powerhouse nations like China, the United States, and even AI-innovative countries such as France. Coupled with the AI initiatives of neighboring UAE, Saudi Arabia's strategy has the potential to establish a new hub of technological influence on the world stage.

However, in Washington, concerns linger over the kingdom's objectives and autocratic tendencies, particularly regarding the potential ramifications for US interests. There is apprehension that Saudi Arabia's collaboration with China, for example, could lead to the provision of computing resources to Chinese entities, potentially undermining US interests.

Saudi Arabia's recent announcement of $1 billion in funding for GAIA, an AI startup accelerator, underscores its determination to attract and retain talent. Mohammed Almazyad, a program manager for GAIA, emphasized the kingdom's desire to diversify its economy away from oil dependency and foster a vibrant tech ecosystem.

Nevertheless, setting up operations in Riyadh presents its own set of challenges, including scorching temperatures exceeding 110 degrees Fahrenheit in the summer and adapting to the religious and cultural norms of a deeply conservative Muslim society. Despite these hurdles, Almazyad remains optimistic about Saudi Arabia's prospects, emphasizing that the kingdom is just beginning its journey into the realm of AI and tech innovation.

China’s Ant, Baidu, Tencent collaborate with US firms OpenAI, Nvidia on publishing first global generative AI standards:

SCMP:? ?Leading Chinese tech companies Ant Group, Baidu, and Tencent Holdings have partnered with global tech giants such as OpenAI, Microsoft, and Nvidia to establish two international standards concerning generative artificial intelligence (GenAI) and large language models (LLMs). These standards, unveiled during a side event at the United Nations Science and Technology Conference in Geneva, Switzerland, were announced by the non-profit World Digital Technology Academy (WDTA).

The released standards include the "Generative AI Application Security Testing and Validation Standard" and the "Large Language Model Security Testing Method." They represent the first global standards specifically addressing GenAI and LLM technologies, pivotal components of increasingly popular AI services like OpenAI's ChatGPT and Microsoft's Copilot, which leverages OpenAI technology. Notably, Chinese search engine giant Baidu has introduced its own AI chatbot, Ernie Bot, while Tencent and Ant have launched their respective LLMs.

While precedents for international standards and regulations on AI existed before the rise of GenAI, notable developments have occurred. In 2021, UNESCO introduced a "Recommendation on the Ethics of AI," adopted by 193 member states. Furthermore, between 2022 and 2023, the International Organization for Standardization (ISO), a Geneva-based non-governmental organization specializing in standards across diverse domains, issued AI-related guidelines covering system management, risk management, and systems utilizing machine learning.

Big Tech

Big Tech keeps spending billions on AI, mostly for new data centers

IBJ: The world's leading tech companies are doubling down on the artificial intelligence (AI) revolution with unprecedented investments, poised to inject tens of billions of dollars into the industry. Google, Microsoft, and Meta have announced substantial increases in their AI-related spending during their recent quarterly earnings calls. Meta, for instance, revised its projected spending for the year by as much as $10 billion, underscoring the scale of these investments.

Google anticipates allocating approximately $12 billion or more each quarter this year towards capital expenditures, with a significant portion earmarked for the establishment of new data centers. Similarly, Microsoft reported a staggering $14 billion expenditure in the last quarter alone, with Chief Financial Officer Amy Hood forecasting further substantial increases in spending.

These substantial investments not only mark some of the largest cash infusions in a specific technology in Silicon Valley's history but also position the major tech firms as central players in the U.S. economy. As other companies, governments, and individual consumers increasingly rely on these tech giants for AI tools and software, their dominance is further solidified.

However, the surge in AI investments is not without consequences. It is also driving up forecasts for energy consumption in the United States. In regions like West Virginia, aging coal plants initially slated for closure are being kept operational to meet the rising energy demands of the burgeoning data center hub in neighboring Virginia.

For deeper insights, refer to the complete article.

Cross-Border Data Transfer – New Measures Clarify Security Review Requirements

China Briefing: China has unveiled new measures outlining the criteria for security reviews concerning cross-border data transfer. This development comes in response to previous legislation mandating companies seeking to export specific data to undergo cybersecurity assessments. The release of these measures addresses the anticipation among foreign firms and large multinational corporations, providing clarity on the regulatory body tasked with conducting security assessments and the procedures necessary for obtaining clearance to transfer data internationally.