Chinese Progress at the Front
AI for DC
Hi
Welcome (back) to The Prompt. AI isn’t well understood, but we learn a lot in our work that can help. In this newsletter, we share some of those learnings with you. If you find them helpful, make sure you’re signed up for the next issue.
[News] Chinese progress at the front
The Chinese Communist Party’s goal of leading the world in AI by 2030 is a multi-front pursuit with an actual, physical front line. While we hear the most about new models, just as significant is CCP headway in getting other governments around the world to adopt its AI. OpenAI’s analysts report notable progress by one start-up, Zhipu AI, that is offering governments what some Chinese media are calling “China’s answer to OpenAI for Countries.” Last week, Zhipu AI Chairman Liu Debing said, “While promoting the development of domestic large-scale model technology, we also hope to contribute China's AI power to the world.”
Zhipu AI (a/k/a Beijing Zhipu Huazhang Technology Co. Ltd.) is one of China’s “AI tigers” – a newer class of billion-dollar foundation-model start-ups representing CCP hopes of building a self-reliant, globally competitive AI ecosystem that rivals America’s and lessens dependence on American tech. Zhipu AI competes with DeepSeek, publicly rejecting collaboration on pre-training, and independently developing rival models with competitive pricing.
Positioning China’s Belt and Road Initiative as the springboard for a multi-pronged “Digital Silk Road” strategy, Zhipu AI is offering infrastructure solutions to governments and state-owned firms across Southeast Asia (Vietnam, Indonesia, Malaysia, Singapore), the Middle East (UAE, Saudi Arabia), and Africa (Kenya), including: 1) sovereign LLM infrastructure; 2) turnkey, private “AI-in-a-box” hardware with Huawei; and, 3) governance know-how.
In practice, this means:
forming an alliance that co-builds national LLMs with ASEAN and other Belt and Road capitals;
opening joint innovation centers;
embedding safety labs such as the new Dubai Content-Safety Lab (corroborated by CN sources); and,
underwriting projects with Gulf capital (the Saudi Prosperity7 Aramco fund participated in a USD $400M round) and Huawei Ascend hardware, albeit not at Stargate-level scale.
The goal is to lock Chinese systems and standards into emerging markets before US or European rivals can, while showcasing a “responsible, transparent and audit-ready” Chinese AI alternative. Our analysts list concrete Zhipu AI initiatives with ASEAN, Indonesia, Vietnam, Malaysia, Kenya, and an unidentified Belt and Road country.
Zhipu Al maintains strong links with China's government and state-owned entities, as evidenced by over $1.4 billion in state-backed investment, numerous government procurement contracts, and active involvement in setting national AI technical standards. Zhipu AI leadership frequently engages with CCP officials, including Premier Li Qiang. It was added to the US Commerce Department’s Entity List in January on the basis that it is advancing the PRC’s military modernization through the development and integration of advanced artificial intelligence research.
As for CCP progress in covert operations, our latest Threat Intelligence Report contains this key takeaway: For now, at least, we appear to be having more impact using AI to investigate China's covert operations than China is in using AI to disrupt.
As we build out our work on OpenAI for Countries, we’ll keep using our vantage point to shed light on CCP progress with governments on the ground.
[Insight] Call us Cautious Optimists
We’ve discussed how decades of Hollywood blockbusters and popular TV shows have portrayed advanced AI as an existential menace, shaping people’s perceptions of the technology through pop culture before AI even got going. Some in our industry are in the business of – and allegedly raising money off of – scaring people without offering solutions. At OpenAI, we have a different view of our shared future with AI. Call it cautious optimism.
As cautious optimists, we’re committed to sharing real information, not yelling fire in the proverbial theater; giving answers that aren’t merely academic; and most importantly, offering solutions without looking to impose them upon others.
Here’s what we think we know about AI and the opportunity it presents:
Technological change historically has created more jobs.
The majority of today’s jobs did not exist before World War II.
The World Economic Forum projects that more jobs will result from AI than existed before it.
Here’s what we don’t know:
Over what timeline will this economic transition occur.
Which sectors will be negatively impacted – as with lamplighters during the scaling of electricity, who will be the impacted sectors of the Intelligence Age.
Who will benefit from the productivity gains.
We’re cautious because we know AI will bring massive change over the next decade, some of which we can foresee and some we can’t. Because we’re fully aware of the power of this technology – hence our incessant red-teaming and other safety/security work. Because while we believe most will use AI for good, we’re not naive about some trying to misuse and abuse the technology. And because we know a world built on democratic AI isn’t a guarantee given China’s ambitions.
We’re optimists because AI is going to grow the economy – even moderate estimates forecast productivity gains that will drive billions of dollars in GDP growth. Because it’s already unlocking new possibilities, especially when it comes to healthcare, science and research. Because we have faith that most people will use AI and our tools to best help themselves, their families, and their businesses. And because we can see a world – within the next few years – where AI is built globally on democratic rails.
We understand why AI can seem intimidating; new technologies always do. But taking a pessimistic view about AI has a real downside. It discourages people from experimenting with AI in their daily lives, denying themselves the benefits that already exist today while leaving them less prepared for future jobs that will increasingly require AI skills and fluency.
OpenAI is motivated by cautious optimism precisely because the stakes are high. By openly acknowledging — and responsibly addressing — the challenges ahead, we can ensure AI scales human ingenuity, powers economic growth, and gives people everywhere the tools to solve hard problems and accomplish things we can’t even imagine today.
[Prompt] Just how hot is it?
Nothing like July heat in June here in DC, with temperatures and the heat index in the high 90s. ChatGPT can help put it into perspective.
[About] OpenAI Forum
Explore past and upcoming programming by and for our community of more than 30,000 AI experts and enthusiasts from across tech, science, medicine, education and government, among other fields.
8:00 PM – 9:50 PM EDT on Jul 24
[Disclosure]
Graphics created by Base Three using ChatGPT.