Hoy, kumusta? Welcome back to Kaisipan!
AI is reshaping classrooms, banks, and lawmaking. The Philippines is right in the middle of it.
Here’s what happened in AI this week:
England's teachers warn AI is eroding students' critical thinking
Philippines still behind on AI adoption despite growing usage
Congress eyes a sharper AI governance framework with 4E Paifa
BSP governor urges banks to go all-in on AI
Google releases Gemma 4, its most capable open model yet
What England's AI-in-Schools Problem Should Tell Us
England's teachers are raising the alarm. And if it's happening there, it's probably happening here too.
The National Education Union in England surveyed 9,000 secondary school teachers. Two-thirds said they've observed a decline in students' core skills in writing, critical thinking, problem-solving, and even basic spelling.
Kids lean on voice-to-text instead of learning to spell. They let AI finish their assignments instead of thinking through them. One English teacher put it plainly, "AI is destroying what learning is."
What's worth noting is how England's government responded. Hindi nila pinigilan ang AI. Instead, the English government announced plans to deploy AI tutoring tools for up to 450,000 disadvantaged students, framing it as a way to give every child access to personalized support.
Hence, the answer to AI misuse isn't a ban. It's education.
That response, though, isn't without its own problems. England's teachers' unions pushed back with 49% opposed the AI tutoring plan, arguing disadvantaged students need human connection, not a chatbot. And 49% of schools in England still have zero AI policy for students or staff. Even with a national curriculum and active education unions, maraming paaralan doon ang nagde-deploy ng AI nang walang malinaw na gabay.
Pano pa kaya tayo?
DepEd has issued AI guidance, but a memo and actual classroom implementation are two very different things. Marami sa ating mga paaralan ang hindi pa stable or wala pang internet, let alone may framework para sa responsible AI use. At ang mga estudyante, ay gumagamit ng ChatGPT with or without anyone teaching them how.
The real lesson here is that proper use of tools are important. Ang problema ay access without education. Kung hindi natin ituturo kung paano gamitin ang AI nang tama, they will use it in a way that they think is right. Iniiwan lang natin silang mag-figure out on their own. Akala nila tama pero in the long run, sila pala ang malulugi.
At sa Pilipinas, wala tayong oras para hintayin pang lumala bago kumilos.
The Philippines Is Using AI But Using It Well Is Another Story
Global AI adoption is growing fast, but the gap between rich and poor countries is widening. A new report found that 24.7 percent of people in wealthier nations used generative AI in the second half of 2025, compared to just 14.1 percent in developing economies. The Philippines sits in that second group — usage grew from 17.1 to 18.3 percent in 2025, steady but slower than advanced economies.
Microsoft Philippines, in a briefing last March, pushed back on the "we're behind" framing. Ang sinabi nila: the Philippines isn't lagging but it has strong potential through a young workforce, high mobile usage, and strong service-sector presence. The ingredients are there. Pero potential lang iyon kung walang intentional na aksyon.
The real constraint, according to both the report and Microsoft, is skills. It's skills. Josh Aquino, Microsoft Philippines' head of communications, said that AI adoption will be faster if people know how to use it. Microsoft has partnered with DepEd to give roughly 25 million students and 1 million teachers access to AI-enabled learning tools.
Pero in fairness, Microsoft is being “proactive” here in the Philippines. Siguro they’re leveraging the advantage sa bansa natin. The Philippines is still a Windows country.
Bes, Ano Yung…AI Ethics Officer
An AI Ethics Officer, o AEO, is a designated role within an organization responsible for ensuring that AI systems are used fairly, transparently, and responsibly.
Kung may AI ang isang company na nag-a-approve ng loans, nag-a-assess ng job applications, o nag-fa-flag ng customer behavior, the AEO makes sure those systems aren't quietly discriminating against anyone.
Companies like IBM, Microsoft, Salesforce, and BCG already have this role, just under different titles like "Head of Responsible AI" or "Global AI Ethicist." Wala pang bansa na nag-mandate nito sa lahat ng companies by law. Kung matuloy ang 4E Paifa, the Philippines could be among the first.
This Is What an AI Bill Should Look Like
Congress has 23 AI-related bills in committee. Daming bida-bida. In our congress na number of bills ang sukatan ng pagiging legislator, hala sige submit lang ng submit. Of that the 23 AI-related bills, House Bill 13 and House Bill 658 are the frontrunners, both proposing to establish a Philippine Council on Artificial Intelligence under DICT, complete with a governance framework, ethics code for developers, and a regulatory sandbox.
But a separate proposal may be the most thoughtful one yet.
AI ethics advocate Dominic Ligot and his team are pushing the Philippine AI Flourishing Act, or 4E Paifa, to the technical working group.
Unlike the two lead bills, 4E Paifa skips creating a new regulatory body. It builds on existing agencies: DOST leads, DICT supports, coordinated through a National AI Council.
The logic: Agencies that already understand health, labor, and finance are better placed to regulate AI in those sectors.
What sets it apart from the current frontrunners:
Mandatory AI Ethics Officer for organizations deploying AI, similar to how the Data Privacy Act normalized the Data Privacy Officer role. Great move!
Codified rights for Filipinos like the right to human oversight of AI decisions, protection from algorithmic discrimination, and the ability to seek redress when AI causes harm
Required Filipino-language AI systems and datasets
Advance notice, free reskilling, and compensation for workers displaced by automation
May tradeoff. 4E Paifa leaves deepfakes to separate legislation, meaning weaker coverage until those laws catch up. Ligot is candid about it. The Philippines is already exposed. But his bet is that a focused proposal that passes beats a comprehensive one that stalls.
Mga Ibang Ganap
Google released Gemma 4, its most capable open-source AI model family yet, with four model sizes that run offline on phones, laptops, and workstations.
Anthropic restricted Claude Code subscribers from using their subscription limits on third-party tools like OpenClaw, requiring a separate pay-as-you-go plan.
Bangko Sentral ng Pilipinas Governor Eli Remolona urged Philippine banks to adopt AI across all operations, acknowledging potential job losses while noting 44% of local banks have already deployed at least one AI model.
The Department of Agriculture in Davao Region partnered with Japanese firm E-Supportlink to pilot AI-powered drones for banana disease detection, covering 15 hectares across Davao del Norte and Davao de Oro.
Ateneo de Manila University hosted researchers from the University of Tokyo's Matsuo-Iwasawa Laboratory to explore AI education partnerships, with a free three-month AI course open to Filipino students starting April 8.
Palaisipan
Is this AI or not?
Prompt Tip!
Ever caught AI confidently citing a study that doesn't exist? It happens more than you think because AI is trained to sound helpful even when it's unsure.
AI Prompt Hackers shared a simple fix for this: before asking your question, tell the AI to rate its confidence on every claim it makes.
Use three levels: HIGH for things it's certain about, MED for things it believes but hasn't verified, and LOW for anything it's inferring or estimating. After it answers, ask it to list all MED and LOW claims separately.
Answer my question below. For every factual claim, do one of three things: 1. Name a specific, real source (publication, study, institution — with year if you know it) 2. Write "(general knowledge)" if it's widely established and you don't have a specific source 3. Write "(inference)" if you're reasoning from related information rather than stating a fact Do not invent sources. If you don't know the source, say so. Question: [YOUR QUESTION]
What you get is a shortlist of exactly what needs fact-checking, instead of reading everything and guessing which parts to trust.
The biggest benefit: you move faster, you share with more confidence, and you stop getting burned by sources that never existed.
That’s all for today!
Palaisipan Answer: The image is AI! 😄
Thank you for reading today’s newsletter!
Do you have suggestions? I’d love to hear your thoughts.