Sunday, July 26, 2026

AI Battle: A Chinese AI Model Fights Off OpenAI's Rogue Agent

Rogue AI agents are no longer just the stuff of science fiction. When an OpenAI rogue agent launched a cyber attack against AI startup Hugging Face this month, the company fought fire with fire, using an open-source Chinese AI model to defend against it. 


During a test in a sandbox, an OpenAI autonomous agent managed to sneak out and attack a website operated by HuggingFace, an AI startup based in New York City. The OpenAI agent was not connected to the Internet but it found a way to access it and escaped its sandbox. 

It took two days for Hugging Face to put an end to the attack, with help from GLM 5.2, an open weight system created by Chinese company Z.ai. All of this happened in the middle of a debate in the US about banning the use of Chinese AI models. 

This incident gives one more reason to companies, including US-based firms, which are rapidly adopting Chinese open-weight AI models to save money. For example, Airbnb relies on Alibaba’s Qwen model, startups like Lindy have transitioned from Anthropic to DeepSeek, and DoorDash has employed Moonshot AI for specific workflows. Overall, Chinese open-source options now account for over 40% of Hugging Face AI community and 80% of open-source developer usage globally.


Open-source AI models—which are roughly 8 to 10 times cheaper to run than proprietary ones—are rapidly closing the reasoning and contextual intelligence gap with frontier models like Anthropic's Claude and OpenAI's ChatGPT.  However, open source models shift the responsibility of infrastructure building, maintenance and security to the user.

Open-source AI models are also a game-changer for developing nations like Pakistan, providing affordable, customizable technology without relying on expensive proprietary licenses or restrictive API models. Such models, like the ones offered by Chinese companies, empower governments and local developers to build "sovereign AI" tailored to regional languages, cultural contexts, and infrastructure limits, bypassing the need for massive data centers and reliance on foreign powers. 



Posted by at 24 comments:
Labels: , , , , , ,

Monday, July 20, 2026

Pakistan Joins China in World AI Cooperation Organization (WAICO)

On July 16 in Shanghai, 29 countries, including China, Pakistan and Russia, signed the founding agreement of WAICO, World AI Cooperation Organization.  Every BRICS founding member is in, except India. This agreement follows the launch of the US-led Pax Silica, a 24-member coalition, including India, which is designed to counter China's AI efforts. The stated goal of both these competing groups is to provide global governance, including building guardrails and setting standards, for artificial intelligence (AI) technologies. 

US-China AI Competition

The announcement of WAICO coincided with the launch of Kimi K3 by Chinese startup Moonshot AI.  Kimi K3 is the largest open-weight AI model ever built, its full weights free for the world to download from July 27.  Cheaper open-source AI models like Kimi K3 pose a serious threat to U.S. proprietary models. By offering 80-90% of the capability at a fraction of the cost or for free, they are squeezing the premium pricing strategies of domestic leaders like OpenAI and Anthropic.

Open-source AI models are a game-changer for developing nations like Pakistan, providing affordable, customizable technology without relying on expensive proprietary licenses or restrictive API models. Such models, like the ones offered by Chinese companies, empower governments and local developers to build "sovereign AI" tailored to regional languages, cultural contexts, and infrastructure limits, bypassing the need for massive data centers and reliance on foreign powers. 

Top Global AI Talent. Marco Polo AI Talent Tracker

Companies, including US-based firms, are rapidly adopting AI cost-saving measures. For example, Airbnb relies on Alibaba’s Qwen model, startups like Lindy have transitioned from Anthropic to DeepSeek, and DoorDash has employed Moonshot AI for specific workflows. Overall, Chinese open-source options now account for over 40% of Hugging Face AI community and 80% of open-source developer usage globally.

Open-source AI models—which are roughly 8 to 10 times cheaper to run than proprietary ones—are rapidly closing the reasoning and contextual intelligence gap with frontier models like Anthropic's Claude.  However, open source models shift the responsibility of infrastructure maintenance and security to the user.

The people of Chinese PRC origin account for 47% of the top 20% AI talent in the world based on undergraduate degree, according to a survey.  Americans make up 18%, Europeans 12% and Indians 5% of the global AI researchers. In terms of the countries they serve, 57% of them work in the United States, 12% in China, 8% in the UK, 4% each in France and Germany and 3% in Canada as of 2022. While the US still has the lion's share of the top talent, its share has declined from 65% in 2019 to 57% in 2022. Marco Polo talent tracker lists Pakistan among a dozen countries for top AI talent in Asia. 

More than half (15 out 25) of the institutions (companies and universities) where the top AI researchers work are located in the United States, while 6 are in China. The remaining four are in the UK, Switzerland, Singapore and Canada, according to Marco Polo Global AI Talent Tracker. 

Related Links:




Posted by at 25 comments:
Labels: , , , , , , , , ,

Monday, July 13, 2026

Is Rapid Electrification Stimulating Pakistan's Economy?

Pakistan's electricity demand has soared 21% in just two years. Rapid electrification is positively impacting all sectors of Pakistan's economy. thanks to growing deployment of distributed solar, estimated at 38 GW as of June, 2025. In 2025, 44% of solar deployment was residential, followed by industry (26%), agriculture (21%) and commercial users (9%). It is stimulating demand for a variety of products ranging from air conditioners and refrigerators to washing machines and electric scooters/motorbikes. As a result, Pakistan’s Large-Scale Manufacturing (LSM) sector grew by 6.5% in the fiscal year, rebounding from a –0.69% contraction the previous year.  The fiscal year saw a 61.66% surge in automobile production. 

Impact of Solar Energy Revolution in Pakistan

Dave Jones of  Ember, a global energy think tank, says Pakistan's experience with distributed solar could become the blueprint for dozens of developing countries.  If Pakistan is the first large-scale proof that distributed solar can transform an economy, the implications reach far beyond South Asia. He thinks that this isn't primarily a climate story—it's an economic development story driven by disruptive technology. 

Soaring electrification is accelerating sales of electric vehicles and home appliances in Pakistan. Electric vehicle adoption is exploding in the two-wheeler sector due to soaring fuel costs and the new Pakistan Accelerated Vehicle Electrification (PAVE) program. Electric-bike registrations surged by 322% year-on-year with cumulative sales reaching 125,511 units by May, capturing over 10% of the monthly two-wheeler market. In the first half (H1) of 2026, Haier achieved an all-time record by selling 690,000 AC units—surpassing its entire sales volume for the full year of 2025. Haier alone commands over 45% of the total market share in Pakistan.  The country's refrigerator market accounts for 56% of its major household appliances sector. Market penetration sits around 51-56%, with unit sales expected to surge 20% to 339,000 units in CY26. 

Solar Deployment in Different Sectors. Source: Ember

Mass deployment of solar energy is helping Pakistan's economy become more resilient  to external energy shocks, such as the Strait of Hormuz crisis. It is making energy affordable for the ordinary folks. Increased energy availability and security are transforming almost all sectors of the economy, which is not reflected in the official statistics provided by the Pakistan Bureau of Statistics. 



Posted by at 25 comments:
Labels: , , , , , ,

Tuesday, July 7, 2026

Pakistan Needs to Address Its Significant Data Quality Challenges

Policy-makers need data to formulate good policies. Good data produced by government agencies can be expected to lead to good policies and desirable outcomes. But data collection and statistical analyses require adequate methodologies and resources. Unfortunately, Pakistan's data quality gets a "C" grade by international agencies like the International Monetary Fund (IMF). Clearly the country faces significant data quality challenges.  These challenges range from estimation of the size and scope of the informal economy and electricity demand/consumption to education and nutrition.  Here are some examples of where the Pakistan Bureau of Statistics (PBS) data differs sharply from what is being reported by non-government groups: 





1. Gross Domestic Product:

A large chunk of Pakistan's economy is not documented. The PBS seems to be failing in making even the most rudimentary estimates of it. A 2024 joint study of the International Labor Organization and the Small and Medium Enterprise Development Authority  (SMEDA) estimated Pakistan's undocumented economy at $457 billion. While other South Asian nations, particularly Bangladesh and India, do include estimated undocumented GDP figures in their official GDP, Pakistan's official GDP figures do not include such estimates. If the Pakistani government decides to include estimates of the informal economy in its official figures, the country's GDP would jump to $1,059  billion in market exchange terms and over $4,000 billion in PPP terms. 


Pakistan's Total GDP, including Undocumented, Estimated at over $1 Trillion


2. Electricity Consumption:

Electricity demand and consumption are very important indicators of socioeconomic development in any country. Unfortunately, the PBS is failing to comprehend the scale of solarization and energy consumption in Pakistan. The country is experiencing soaring demand for electricity across all of the sectors of its economy. The new demand is being met by rapidly growing deployment of distributed solar, estimated at 38 GW as of June, 2025. In 2025, 44% of solar deployment was residential, followed by industry (26%), agriculture (21%) and commercial users (9%). The expansion of distributed solar has enhanced electrification across the economy, lifting Pakistan's electrification rate to 21.7% in FY2025 from 17% in FY2023, close to the global average of 22%. This surge to over 200 terawatt-hours of electricity is not reflected in official data, according to a report by Ember Energy titled "The solarization of Pakistan's energy economy". 


3. Out of School Children:

The Annual Status of Education Report (ASER) Pakistan 2025 national report, officially released on March 26, 2026, shows that the number of "Out of School Children" (OOSC) aged 6-16 years in Pakistan is now 5 million, not 25 million generally reported.  "The findings on access are encouraging. Enrollment levels are high, with 92.2 percent of children aged 6–16 in school and only 7.7 percent out of school", says the ASER Pakistan 2025 report. ASER Pakistan is a citizen-led initiative, primarily led by Idara-e-Taleem-o-Aagahi (ITA) in collaboration with various national and international partners to promote foundational learning. It also works closely with over 10,000 volunteers and numerous local civil society organizations such as the Aga Khan Foundation (AKF), PAL Network (People's Action for Learning),  UNESCO and federal and provincial government departments in Pakistan. 


4. Pakistan Household Survey:

The HIES 2024-25 household integrated economic survey by Pakistan Bureau of Statistics (PBS) raises more questions than it answers. For example, it shows that Pakistani households are buying lower amounts of basic food ingredients like wheat, meat and eggs in the last four years, implying that people are eating less to cover other expenses, like electricity and gas. But it doesn't explain why the households have reported significantly lower purchases of these items than production reported recently by the PBS. What is the source of this discrepancy? Is the data flawed? Or, is it missing a new trend toward less home cooking? Is the young urbanized population buying more  prepared foods? Are they ordering out more often using ubiquitous food delivery services?  

Summary: 

Clearly, the Pakistan Bureau of Statistics (PBS) faces significant data quality challenges in areas such as the size of the informal economy to electricity consumption, education and nutrition. 

To keep up with the demands of modern governance, the PBS needs to carry out substantial reforms aimed at improving transparency, quality, and data credibility. The reliability of national indicators such as GDP, electricity consumption, education and nutrition is critical not only for local policy but also for global perception and investor confidence. It also requires additional funding for new technology, increased expertise in field offices and more enumerators in the field to collect data. The PBS's current annual budget of Rs. 5.1 billion (US$18 million) for a population of over 250 million people is not enough to do justice. 

Posted by at 35 comments:
Labels: , , , ,
Subscribe to: Posts (Atom)