I’ve been sitting with a question that’s been rattling around in my head: If AI doesn’t feel pain and frustration like human software engineers do, would it innovate? Would it address pain points it doesn’t encounter?

This episode came together after I posed this thought experiment on social media a few weeks back, and I got some genuinely fascinating responses from the community. Would AI have invented jQuery? React? TypeScript? Would it have even created human-readable programming languages? These questions led to a really rich discussion—one I wanted to dig deeper into here.

I also share some personal updates about finishing the academic portion of my master’s program and starting my dissertation journey in September. But the meat of this episode is exploring what several thoughtful commenters had to say, including Geoffrey Hinton’s perspective on digital intelligence and generalization, and the really compelling example of AlphaZero reinventing 3,000 years of Go strategy.

Here’s what I ultimately found: the answer might hinge on how we define innovation itself, and whether discovering new mathematical paths to existing goals counts as true invention or something else entirely.

Timestamps:

- 00:15 – Master’s program updates

- 04:18 – The Thought Experiment and community responses

- 06:13 – John Kendrick’s take on Geoffrey Hinton and AlphaZero

- 09:17 – My own analysis and thoughts

- 14:00 - Is discovery the same as innovation?

- 16:04 – Final reflections

I’d love to hear what you think. Is AI capable of true innovation, or are we looking at sophisticated pattern matching? Drop your thoughts in the comments.

Thanks for tuning in to the AI Guardrail podcast as always!

See you soon,

Rem (Kuya Dev)

Host, AI Guardrail Podcast

Hello there!

Today, I will go deep, very deep into the current state of upcoming AI regulatory bills in the Philippine Congress as compared to the EU AI Act. This is part of a research paper I did for my AI Ethics class in graduate school. So brew a cup of coffee, take a seat, get comfy, and enjoy the ride.

Introduction

On August 1, 2024, the EU AI Act was enacted after years of deliberation. It is the first ever regulation on AI in the world, and may soon become the benchmark for global AI regulation efforts, following the footsteps of GDPR, which is among the most influential data protection laws in the world (NetSuite.com, 2024). It is a risk-based statute, built on the foundations of European values and ethics, seeking to position the region to be a global leader in AI innovation without harming the fundamental rights and well-being of Union citizens. Several countries are now looking to it as a possible blueprint for their own AI regulatory policies, including the Philippines.

In this post, we look back at the evolution of the EU AI Act, and examine its parallels and contradictions with a few of the AI regulatory proposals filed in the Philippines House of Representatives and its applicability in the broader Philippine context.

EU AI Act: From Conception to Enactment

The EU AI Act didn’t come into fruition overnight. It was a result of almost a decade of evolution that is grounded on Europe’s values- and rights-first culture and regulatory approach. (Bradford, 2023)

Before AI automation came into the public consciousness, ethical standards on robotics automation first took center stage in Europe, led by British Standard BS 8611 in 2016. In the same year, the IEEE Standards Association launched a global initiative on the Ethics of Autonomous and Intelligent Systems, with the purpose of drafting standards that will steer these technologies into becoming human-centered rather than an exclusively profit-driven endeavors, and effectively generalizing system autonomy to include intelligent systems beyond robotics. (Directorate General for Parliamentary Research Services, European Parliament, 2020)

On May 25, 2018, GDPR was enacted, setting the precedent for other successive technology legislation, including the AI Act, as it established that the EU is capable of regulating fast-changing digital technologies. It also extended the “Brussels Effect”, which recognizes the heavy influence of EU policies on global regulatory efforts owing to its unique geopolitical makeup (Bradford, 2012), to the digital space.

One of the major milestones is the Communication from the Commission to the European Parliament in 2018 which set the direction of where the AI legal and regulatory conversations should be heading. It argued that AI as a driver of economic growth should not be exclusive to maintaining alignment with European values, rights, and ethical standards.

In April 2019, The Ethics Guidelines for Trustworthy AI was published by the High-Level Expert Group on AI, convened by the European Commission. This established the common vocabulary for what “trustworthy AI” means, and three components that it should meet throughout its life cycle: lawful, ethical, and robust. These components were further broken down into seven core principles: human agency and oversight; technical robustness and safety; privacy and data governance; transparency; diversity, non-discrimination, and fairness; societal and environmental well-being; and accountability. It invited stakeholders to voluntarily integrate its proposed AI ethics assessment list into their systems.

The European Commission published The White Paper on AI in 2020 sounding the alarm on the dangers of a fragmented approach of EU member states to AI and pushed for a harmonized policy. This was also the beginnings of the risk-based approach that the EU AI Act is now well-known for, and proposed for stricter guidelines and obligations, while balancing excellence (innovation) with trust (rights preservation through AI regulation).

The first draft proposal for a concrete regulatory framework on AI was finally submitted by the Commission on April 21, 2021. This was groundbreaking, as it was the first ever comprehensive AI legal regulatory document in the world. It was cross-sector, and integrated the risk-based framework that the 2020 White Paper on AI suggested. It also framed AI as a product, which will thereby enable applicability of already existing mechanisms for product market surveillance and regulations.

What followed is a series of dialogues between the EU Council and Parliament. The former crafted their position in November 2022, with the latter releasing theirs in June 2023. A trilogue together with the EU Commission commenced soon after. The major contentions are listed in the table below.

The trilogues happened around the time LLMs like ChatGPT began to gain public adoption (OpenAI, 2022). It forced the policy makers to immediately include General Purpose AI (GPAI and Generative AI (GenAI) into the drafts for the act, with the Parliament’s position document explicitly mentioning ChatGPT in the text. A provisional agreement was finally reached in September 2023 after a “3-day marathon talks” (European Council, 2023), adopting the following changes:

• Rules on high risk and high impact GPAI

• A new governance architecture which would consist of an AI Office under the EU Commission, a panel of independent experts that would serve as GPAI advisors to the AI office, an AI Board under the EU Council, and an advisory forum of stakeholders

• Prohibition list expansion, with exemptions for remote biometric identification by law enforcement in strict situations with safeguards

• Pre-deployment impact assessment obligations of high-risk AI systems deployers

The AI Act was signed into law on June 13, 2024, published in the official journal a month after, and enacted on August 1, 2024. with its implementation to be conducted in phases. This sets the stage for the whole world to follow suit, as the Brussels effect is set in motion.

The EU AI Act as Benchmark for Philippine AI Regulation

The Philippines is racing to adopt its own AI regulatory legislation. In fact, 19 bills have been filed in the Congress just in the second half this year alone, not only reflecting the pressing need for a national AI policy, but also of the fragmented nature of local politics in the country. Whereas the EU brought forth its AI Act through collective effort, partly due to fears of fragmentation, the Philippine approach is grounded in individualism, personality politics, and optics (Leon, 2025). As evidenced by the surnames of some of the bill sponsors, powerful and influential political dynasties (Panti, 2025) are once again attempting to embed themselves in what could be one of the landmark legislations in the country. Political agenda aside, though, the large number of proposals signals that the Lower House is in the midst of recognizing and addressing the urgency posed by the lack of protections against potential misuse and abuse of AI.

The 19 house bills are presented in the succeeding table. For the rest of the section, we take on the challenge of consolidating the provisions in all of the bills, and attempt to perform a comparative analysis with the EU AI Act to determine the extent of its influence, and to spot stipulations that are unique to the Philippine context. Note that HB0013, HB00658, HB01920, HB02827, HB 03195, and HB05158 are all mirrors of each other, i.e. exactly the same bills but with different sponsors. Meanwhile, HB00073, HB00252, and HB03688 share the same title but only HB00073 and HB03688 are mirror bills. This means there’s effectively just 13 house bills, fewer albeit still considerable. To introduce a bit of clarity, the mirror bills are marked accordingly in the table. Additionally, for the rest of the text, we will only refer to the first bill (highlighted in boldface) to represent the group, and disregard its mirror filings.

Core Philosophies

The EU AI Act’s goal is to harmonize rules across member states while protecting fundamental rights. It relies on already existing legislation to provide mechanisms to protect human rights and welfare, while imposing the following obligations on providers and deployers of high risk AI systems to ensure safety, fairness, and accountability:

• Risk Management Systems

• Data Governance

• Transparency and Human Oversight

• Robustness, Accuracy, and Cybersecurity

On the other hand, a couple of Philippine proposals, HB00013 and HB00057, are aiming to leverage AI for national development while adopting rights protections through a more explicit “Bill of Rights”, seeking to empower citizens with the power to demand protection of their rights and to seek legal remedies whenever they are violated:

HB00013 Bill of Rights

• Right to Protection from Unsafe and Ineffective AI Systems

• Right Against Algorithmic Discrimination

• Right to Privacy (compliance with Republic Act No. 10173, “Data Privacy Act of 2012”)

• Right to Know

• Right to Remedy

HB00057 Bill of Rights

• Right to Human Agency

• Right to Algorithmic Transparency

• Right to Protection from Unjust Harm

• Right to Data Sovereignty

• Right to Safe and Accountable Systems

• Right to Contestability and Redress

• Right to Non-Obselescense

• Right to Participate in Governance

Risk-Based AI Classification

A couple of the legal proposals in the Philippines for an AI regulatory framework seek to adopt a risk-based AI system classification, evidently influenced by the EU AI Act.

Although HB06028 only explicitly stated High, Moderate, and Low Risk, its risk tiers align with the High, Limited, and Minimal tiers of the EU AI Act. HB01196, meanwhile, is almost an exact replica of its EU counterpart.

AI Prohibitions

Zooming into the AI Prohibitions allow us to view key differences between the approaches, which is summarized in the succeeding table.

There is significant overlap on most of the harm categories. Glaring differences are the explicit inclusion of the ban of lethal autonomous weapon systems (LAWS) as emphasized multiple times in HB00013, and a strict stance on deep fake and misleading technologies in HB00057, HB00252, HB03480, HB03905, and HB06028 as opposed to softer transparency mandate in the EU AI Act.

Governance and Enforcement

The EU AI Act adopts a multilayered governance architecture by creating multiple regulatory bodies already mentioned earlier: the EU AI Office, European AI Board, Board of Scientific Experts, and Competent National-Level Authorities.

Meanwhile, several of the Philippine bills propose the adoption of a multi-agency approach, leveraging already existing government organizations through cross-functional collaboration. The proposed structure will include the following:

• Philippine Council of Artificial Intelligence (PCAI) - multi-sectoral experts and advisers in charge of high level policy creation and will serve under the Department of Information and Communications Technology (DICT)

• Artificial Intelligence Board (AIB) - primary regulatory and supervisory body composed of high-ranking government officials from different government organizations such as the DICT, Department of Science and Technology (DOST), Department of Trade and Industry (DTI), Intellectual Property Office of the Philippines (IPOPHIL), National Privacy Commission (NPC), and National Intelligence Coordinating Agency (NICA)

• Artificial Intelligence Development Authority (AIDA) - alternative to AIDA, under the Department of Science and Technology (DOST)

• National Center for AI Research (NCAIR) - primary body that will be tasked with advancing AI innovation and developing AI talent in the Philippines, and will be placed under either DOST or DICT

Penalties

The EU AI Act implements a tiered system of hefty fines:

• Prohibited AI Practices - up to €35 million or 7% of total worldwide annual turnover

• Non-Compliance - up to €15 million or 3% of global turnover

• Inaccurate Information - up to €7.5 million or 1% of global turnover

The Philippine House Bills combine both fines and criminal liability that may lead to imprisonment:

• Fines - up to ₱ 10,000,000

• Imprisonment - from six months up to six years

Startups and SME Support

Both the EU and the Philippines recognize the need to allow small businesses and startups to flourish, and have included provisions that intend to provide space and flexibility for AI startups and SMEs in the form of regulatory sandboxes.

Regular fines in the EU require the business to pay the higher of the two options (hard cap or percentage cap) but small businesses may pay the lower of the two instead.

Some of the Philippine bills are seeking to provide funding and incentives to encourage building of AI businesses and ventures.

Protections Against Job Displacements

The EU AI regulation delegates worker protections to existing laws on social policy and labor. The Philippine Bills, in contrast, explicitly mandate a Job Displacement Program to be managed by the Department of Labor and Employment (DOLE) for job losses due to AI. Both require the disclosure of usage of AI to employees especially in high risk applications, and the need of human oversight to minimize risks. Employee training and reskilling on AI usage and development are also advised.

Summary

The EU AI Act extends on the legacy of the GDPR in setting the gold standard of technology legislation. It introduced a novel risk-based approach to AI governance and human rights protections. From surface level, it may seem like the EU AI Act is very lax in its enforcement of rights protections, e.g. no provisions for imprisonment, one must also consider that the law wasn’t designed to replace any criminal law that’s already existing (European Commission, 2021), and it actually relies on them to provide the legal framework for any violations that endanger human life and well-being.

While there are obvious inspirations taken by some of the Philippine bills from the EU AI Act, there are also efforts to differentiate the Philippine context from the EU experience. The political nature of the Philippines has resulted in a fragmented approach to drafting a national AI governance policy, making this particular comparative study extremely challenging to accomplish even with the presence of LLMs that aid in making sense of all the documents. I do recognize, though, that had it not been for LLMs, this would not have been possible. It is also worth noting that this paper’s scope was confined to bills filed in the Philippine House of Representatives, and deliberately left out similar initiatives in the Philippine Senate.

The Philippines is still in its early stages of laying down the groundwork for a national AI law. Hearings on these bills are ongoing as I write this, and they will need to overcome both legal and political hurdles before they can all be consolidated into a unified initiative. The president has not even included the need for an AI law in his list of priority bills, choosing to shift the government’s focus instead to anti-corruption measures, and for good reason (Mangaluz, 2025). As a citizen and technology professional myself, I can only hope that the Philippine AI Act that finally comes out of this will be one that serves the interests of the people, not of its politicians nor of the huge conglomerates that dominate Philippine society, and that its enforcement will not go the route of the many laws in the country: poor to no enforcement at all (Philipp, 2023; The Philippine Star, 2021).

Acknowledgement

I would like to express my gratitude to AI ethicist and friend Doc Ligot, for making these bills available publicly through his personal Google Drive. These legal documents would most likely not have been accessible without his efforts.

AI Usage Disclosure

In order to augment my lack of legal training and skills, I utilized LLM technologies. ChatGPT was used in order to gain better understanding about the evolution of the EU AI Act. Meanwhile, to consolidate and synthesize the 19 bills currently filed in the Philippine Congress, and conduct a comparative analysis with the EU AI Act, I turned to Google’s NotebookLM.

However, while content in this paper is the result of interactions with ChatGPT and NotebookLM and may have been inspired by them, none of the material has been directly generated by AI. Additionally, I had to perform due diligence in making sure that the output is grounded on the source materials, and not from AI hallucination.

Sources

• An Act Establishing a Comprehensive Legal Framework for the Development, Deployment, and Regulation of Artificial Intelligence in the Philippines, Creating the Artificial Intelligence Development Authority (AIDA), and for Other Purposes, No. 1196, House of Representatives (2025).

• An Act Establishing a Legal Framework for the Responsible and Accountable Use of Artificial Intelligence in the Philippines, No. 3480, House of Representatives (2025).

• An Act Establishing a Regulatory Framework for a Robust, Reliable, and Trustworthy Development, Application, and Use of Artificial Intelligence Systems, Creating the Philippine Council on Artificial, Delineating the Roles of Various Government Agencies, Defining and Penalizing Certain Prohibited Acts, 13, House of Representatives 1 (2025).

• An Act Instituting the National Artificial Intelligence Code of the Philippines, Creating the Bureau of Artificial Intelligence Systems Under the Department of Information and Communications Technology, and Establishing a Comprehensive Framework for the Safe Ethical and Inclusive Governance of Artificial Intelligence Systems in the Country, and for Other Purposes, 57, House of Representatives 1 (2025).

• An Act Mainstreaming Artificial Intelligence Into Government Policy Formulations, Establishing the Framework Strategy and Program on Artificial Intelligence, Creating for This Purpose the Artificial Intelligence Commission, and for Other Purposes, No. 2323 (2025).

• An Act Mandating the Department of Education, Commission of Higher Education, Department of Science and Technology and Department of Information and Communications Technology to Develop Generative Artificial Intelligence (GenAI) Regulations to Maximize Its Positive Impact on Philippine Education System, 598, House of Representatives 1 (2025).

• An Act Promoting Data-Driven Governance and Artificial Intelligence Solutions in the Public Sector Through Government-Industry-Academia Collaboration, and Appropriating Funds Therefor, 1484, House of Representatives 1 (2025).

• An Act Promoting the Development and Regulation of Artificial Intelligence in the Philippines, 73, House of Representatives 1 (2025).

• An Act Promoting the Development and Regulation of Artificial Intelligence in the Philippines, 252, House of Representatives 1 (2025).

• An Act Promoting the Development and Regulation of Artificial Intelligence in the Philippines, Creating for the Purpose the Philippine Artificial Intelligence Commission, and for Other Purposes, No. 3462, House of Representatives (2025).

• An Act Providing a National Strategy for the Development of Artificial Intelligence, Creating for the Purpose and National Center for Artificial Intelligence Research, and Appropriating Funds Therefor, 659, House of Representatives 1 (2025).

• An Act Recognizing the Use of Artificial Intelligence and Generative Technologies in the Commission of a Crime as an Aggravating Circumstance, and Providing for the Liability of Developers and Platforms Thereof, 3905, House of Representatives 1 (2025).

• An Act Regulating the Development and Use of Artificial Intelligence Systems in the Philippines, Promoting Ethical and Responsible Artificial Intelligence Innovation, and Integrating Sustainability and Futures Thinking in National Policy Making, and for Other Purposes, 6028, House of Representatives 1 (2025).

• Bradford, A. (2012). The Brussels Effect. In A. Bradford, The Brussels Effect (1st ed., pp. 25–66). Oxford University PressNew York. https://doi.org/10.1093/oso/9780190088583.003.0003

• Bradford, A. (2023). The European Rights-Driven Regulatory Model. In A. Bradford (Ed.), Digital Empires: The Global Battle to Regulate Technology (p. 0). Oxford University Press. https://doi.org/10.1093/oso/9780197649268.003.0004

• Communication From the Commission to the European Parliament, the European Council, the Council, the European Economic and Social Committee and the Committee of the Regions - Artificial Intelligence for Europe (2018). https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52018DC0237

• Directorate General for Parliamentary Research Services, European Parliament. (2020). The ethics of artificial intelligence: Issues and initiatives. Publications Office. https://data.europa.eu/doi/10.2861/6644

• European Council. (2022, November 25). Proposal for a Regulation of the European Parliament and of the Council laying down harmonised rules on artificial intelligence (Artificial Intelligence Act) and amending certain Union legislative acts—General approach. Consilium. https://data.consilium.europa.eu/doc/document/ST-14954-2022-INIT/en/pdf

• European Council. (2023). Artificial Intelligence Act: Council and Parliament strike a deal on the first rules for AI in the world.

• European Parliament. (2023). Parliament’s negotiating position on the artificial intelligence act.

• High-Level Expert Group on AI. (2019, April 8). Ethics guidelines for trustworthy AI | Shaping Europe’s digital future. https://digital-strategy.ec.europa.eu/en/library/ethics-guidelines-trustworthy-ai

• Leon, D. de. (2025, June 30). Your lawmaker filed either hundreds, or just a few bills. So what? RAPPLER. https://www.rappler.com/newsbreak/explainers/context-nuances-how-many-bills-resolutions-filed-congressman/

• Mangaluz, J. (2025, December 9). Marcos to Congress: Prioritize Anti-dynasty bill, other transparency measures. Philstar.Com. https://www.philstar.com/headlines/2025/12/09/2493003/marcos-congress-prioritize-anti-dynasty-bill-other-transparency-measures

• NetSuite.com. (2024, December 4). What Is General Data Protection Regulation (GDPR)? Oracle NetSuite. https://www.netsuite.com/portal/resource/articles/business-strategy/gdpr.shtml

• OpenAI. (2022, November 30). Introducing ChatGPT. https://openai.com/index/chatgpt/

• Panti, L. (2025, December 10). Anti-political dynasty bill: Ever present and absent at the same time. GMA News Online. https://www.gmanetwork.com/news/topstories/nation/969040/anti-political-dynasty-bill-ph-timeline/story/

• Philipp, J. (2023, April 3). Strengthening the Rule of Law in the Philippines. The Borgen Project. https://borgenproject.org/rule-of-law-in-the-philippines/

• Proposal for a Regulation of the European Parliament and of the Council Laying Down Harmonised Rules on Artificial Intelligence (Artificial Intelligence Act) and Amending Certain Union Legislative Acts (2021). https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:52021PC0206

• Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the Protection of Natural Persons with Regard to the Processing of Personal Data and on the Free Movement of Such Data, and Repealing Directive 95/46/EC (General Data Protection Regulation) (Text with EEA Relevance), 119 OJ L (2016). http://data.europa.eu/eli/reg/2016/679/oj

• Regulation (EU) 2024/1689: Harmonised Rules on Artificial Intelligence (2024). http://data.europa.eu/eli/reg/2024/1689/oj

• The Philippine Star. (2021, October 17). EDITORIAL - Weak rule of law. Philstar.Com. https://www.philstar.com/opinion/2021/10/17/2134647/editorial-weak-rule-law

• White Paper on Artificial Intelligence - A European Approach to Excellence and Trust (2020). https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52020DC0065

Hello there!

WARNING: This is a long one, but I’ll connect it to AI at the end, I promise!

The other day, I posted on Facebook complaining about this month’s high electricity bill. My wife and I are finding it funny that everyone is now lecturing us on household energy consumption because of it. 🤣

Totally my bad, I should’ve provided more context. First of all, I hate repeating myself, since my career background is a story I've already told soooo many times before. So much so that I didn't think I needed to tell the story again because I feel people are getting tired of it. But I forgot that not all people have heard of it, especially new followers and non-followers of the page. I shouldn't have assumed everyone already knew. 😅 But then again, I also didn’t expect my post to blow up, since it’s not the usual content I put out. I just wanted to vent my anger, man. But alas, here we are.

If you've heard my story, feel free to ignore this if you want. Although this time I'll go a bit deeper. It's been years since we both left Meralco, the largest power distribution utility in the country, so I guess it's safe to divulge more details about our previous jobs there. So here goes...

Guys, my wife and I are former Meralco employees. (PLOT TWIST!) It was both our first jobs fresh out of college, having graduated with an economics and an electrical engineering degree, respectively. We’ve had lengthy careers at Meralco. After almost a decade, I left in 2016 having successfully switched careers into tech. My wife, meanwhile, left in 2019 completing seven years of service.

We were not SIMPLY Meralco employees either, as we were assigned in departments that worked directly on energy sourcing and dispatch, and the operation and monitoring of the WHOLE energy distribution system. I won't go too much into details, but we had direct knowledge of where energy was being sourced, how it was transmitted from the power plants to the Meralco franchise up to the substations and, finally, to the distribution transformers perched on poles that serve your areas. We had direct contact with the power generation companies Meralco had contracts with, WESM, and of course, NGCP. As in one phone call away. 🤣 With a single click of a mouse, we were capable of cutting off and restoring power in any area of the franchise, from streets to blocks to barangays to whole cities. We were also in charge of dispatching and coordinating field crews during maintenance and emergency power restoration efforts. Looking back, it was a pretty cool gig, salary notwithstanding.

[Fun facts: I even ran an anonymous blog in the late 2000s to early 2010s where I occasionally provided information and explainers about scheduled power interruptions, storm-induced blackouts, and the energy crisis we had back then. Meanwhile, my wife once graced the cover of the monthly company magazine (she’s still embarrassed about it lol). We also both won medals in different sports categories at the MVP Olympics, an annual sports event held between companies under the Metro Pacific umbrella, which also included PLDT, Smart, TV5, Maynilad, etc.]

Why am I bringing these all up? To gloat? No, not at all. I just want to put things into perspective before I say this: most of you missed the point of the original post. Again, it’s totally my fault for not framing things properly, not yours. Nevertheless, I sincerely thank all the people who shared their “energy saving” tips in the comments. I’m pretty sure a lot of people learned from them. But yeah, that was not the point of the post.

First, I may not know everything, but my background in electrical engineering and the power industry should at least count for something. It equips me with a particular set of skills that enables me to critically examine our electric bill. My rant had nothing to do with broken meters, nor jumper cables from unscrupulous neighbors, nor appliance efficiency, nor can it be solved by solar panels, especially since we live in a condominium unit.

Second, the pattern of our consumption history doesn’t indicate anything out of the ordinary. In fact, our total kWh consumption was actually LOWER than last year! This meant that our energy conservation efforts were working!

Lastly, the point of the original post was that despite our consumption being lower year-on-year, despite us using the AC much less than last year, despite my wife not working from home anymore… The Peso value on our electricity bill still went up significantly! Comparing with July which had the same level of consumption as last month, our bill today is higher by 12.5%!

Look at it this way, for every PHP 100 we paid last July, we would now have to pay PHP 12.5 more for the exact same kWh consumption. For every PHP 1,000 paid last July, we would now have to pay an additional PHP 125. Our actual bill last July 2025 was around PHP 5,600, now it’s PHP 6,300. Again, FOR THE EXACT SAME KWH CONSUMPTION.

Reiterating what I said in the post, a 12.5% increase in less than a year is absurd. I’m surprised that no one, especially the government, is alarmed about this. Meralco has been steadily increasing their rates these past few months. Recently, the ERC approved another FOUR BILLION PESO recovery package for Meralco. With generation costs expecting to increase due to the stupid war in the Persian Gulf, and (I promised I’d make the AI connection) the government’s push for more energy-intensive data centers in the country, things are just going to get worse, way worse. (I would’ve been less annoyed if I knew that all these profits were shared fairly with Meralco employees but that’s a topic for another day.)

Yes, we absolutely should be doing our own part in reducing our household energy consumption. But the government should also heavily scrutinize distribution utilities (like Meralco), NGCP, and the power generation companies. Maybe it’s high time we revisit EPIRA law, which has disastrously failed its goal of lowering energy prices, and even reconsider nuclear energy. We already have the third highest electricity rates in the whole of Asia, and from the looks of it, we’re gunning for the top spot.

What do you think? Is Meralco getting away with robbery? What’s the situation in your country? Let me know on the comments!

Thanks for reading all the way through. Until next time!

See you soon,

Rem (Kuya Dev)

Host, AI Guardrail Podcast

I’m back after a month away from the podcast, as life got pretty hectic lately. But I’m excited to share what I’ve been thinking about, especially regarding a topic that sparked some interesting conversations on social media.

In this episode, I dive deep into a claim that’s been on my mind: vibe coding is fundamentally probabilistic, while traditional programming is deterministic. And no, before you think I’m against AI-assisted development—I’m not. What I want to do is help you understand the crucial difference between these approaches so you can make informed decisions about how to use them responsibly.

We’ll walk through what probabilistic and deterministic systems actually mean, examine how they apply to both vibe coding and manual code writing, and explore what that trade-off really costs you. I also share some personal updates about streamlining my machine learning workflow for graduate school.

Whether you agree with my take or not, I’d love to hear your thoughts. Comment them down below!

Episode Timestamps:

- 0:00 — Introduction and update on my MSc in Responsible AI journey

- 6:55 — Main topic introduction: Vibe coding vs. traditional programming

- 8:08 — Defining probabilistic vs. deterministic systems

- 10:59 — How this applies to vibe coding

- 12:31 — How this applies to traditional programming

- 16:40 — Why people think traditional programming is probabilistic

- 20:24 — The accountability question and trade-offs

- 21:14 — Wrapping up the core argument

Thanks so much for tuning in—it means a lot to have you here. If this resonates with you, I’d really appreciate a like or share to help this conversation reach more people who care about responsible AI development.

Looking forward to hearing your perspective in the comments, and I’ll catch you on the next episode.

See you soon,

Rem (Kuya Dev)

Host, AI Guardrail Podcast

I’m thrilled to share that I’ve officially become a certified AI Governance Professional (AIGP)!

In this episode, I walk you through my entire certification journey: why I pursued it, how I prepared, what makes the exam so challenging, and concrete tips if you’re thinking about taking it yourself.

Beyond the certification, I also dive into updates from my master’s program in responsible AI—including a fascinating session on real-time language translation from a Google engineer—and share my critical take on the latest open-source AI agent hype (Moltbot/OpenClaw).

Episode Highlights & Timestamps:

- 0:00 – Introduction and my AIGP certification announcement

- 7:20 – Graduate school updates: Computer vision, NLP, and learning from industry experts

- 12:00 – The Moltbot phenomenon: Why the hype doesn’t match reality

- 17:30 – Why I pursued AIGP certification (and why it matters beyond lawyers)

- 28:30 – The exam structure, difficulty, and why practice exams are essential

- 32:00 – Does the certification make you an expert?

- 35:40 – My top tips for aspiring AIGP candidates

- 40:30 – Who needs this certification and why governance bridges the gap between tech and law

Like I promised here are resources to help you (these are NOT affiliate links):

• IAPP AIGP Certification (Purchase)

• AIGP Body of Knowledge (version 2.1, 2026)

• AIGP Study Guide

• IAPP Memberships

• Official AIGP Practice Exam

• Dr. Kyle’s AIGP Masterclass on his website

• Dr. Kyle’s AIGP Masterclass on Udemy (just purchase it from his website instead)

• Official IAPP AIGP Online Training

AI governance isn’t just for lawyers—it’s for engineers, product managers, and anyone building AI systems. This certification grounded me in the legal frameworks that actually constrain what’s possible, and that’s invaluable knowledge for anyone serious about responsible AI.

If you’re exploring AI, AI governance or AI ethics professionally or out of curiosity, I genuinely believe the AIGP path is also worth considering.

Thanks for being part of the AI Guardrail community. Your support means everything.

See you in the next one,

Rem “Kuya Dev” Lampa

P.S. If you have questions about the certification or want to discuss responsible AI, drop a comment or reach out. I’m always happy to talk shop.

---

Like what you heard? Subscribe, share, and support the conversation around ethical AI.

I’m diving into three significant actions the Philippine Department of Information and Communications Technology (DICT) has taken recently, and I wanted to unpack them with you in this episode.

These aren’t simple tech decisions—they touch on governance, privacy, free speech, and the responsibility that comes with government technology initiatives.

What we discuss:

The conversation covers three main areas: the DICT’s exaggerated claims about blockchain security for the national budget, the decision to ban Grok, and a new proposal requiring social media verification for Filipino citizens. I’ll walk through why I believe some of these approaches miss the mark, not because they’re trying to solve the wrong problems, but because the solutions themselves carry risks we need to think through carefully.

I also share a bit about my current journey in my Master’s program in Responsible AI and my preparation for the AI Governance Professional exam.

Episode Timestamps:

- 00:00 – Introduction and episode overview

- 01:44 – Personal introduction and career transition story

- 05:02 – Updates on my Master’s in Responsible AI

- 08:36 – Main topic: DICT’s recent actions begin

- 09:14 – The blockchain press conference and the 101% unhackable claim

- 19:40 – What responsible communication should sound like

- 33:00 – The Grok ban and questions about implementation

- 43:13 – Social media verification proposal and privacy concerns

- 48:52 – The right to voice concerns (even without perfect solutions)

- 52:00 – Closing thoughts and call to action

Thank you so much for taking the time to listen. I know this episode ran long and touched on heavy topics—that’s partly because these issues matter, and partly because the DICT’s recent decisions have sparked important conversations across our tech community.

I don’t expect everyone to agree with me, and honestly, you don’t have to. What matters most is that we’re all thinking critically about how technology shapes our governance, our privacy, and our society.

If you found value in this episode, I’d genuinely appreciate it if you’d share it with others, leave a comment with your thoughts, or subscribe wherever you listen to podcasts—whether that’s YouTube, Spotify, Apple Podcasts, or here on Substack.

Looking forward to continuing this conversation with you in the next episode.

Take care, and I’ll see you next time.

—Rem “Kuya Dev” Lampa

In this episode, I discuss what motivated me to launch this podcast, tracing it back to a TEDx talk I gave in Davao where I first encountered the concept of “data colonialism.” I explore why I felt compelled to become a voice for responsible AI, particularly within the global south where these conversations are still largely absent.

The episode also features my analysis of recent remarks by AMD CEO Lisa Su about scaling AI compute 100x and her projection that 5 billion people will be using AI by 2031. I examine what these ambitious claims mean for populations in developing nations and question whether such projections account for the billions living below the poverty line. Here’s the link to her CES interview.

Throughout, I share my perspective on why we must push back against narratives that treat endless AI scaling as inevitable, and why small businesses and communities shouldn’t feel pressured to adopt technology simply for the sake of growth.

Timestamps

- 00:02 — Introduction and episode overview

- 02:34 — My TEDx talk and the genesis of this project

- 08:22 — Discovering data colonialism and AI’s hidden costs

- 14:50 — Why I’m starting this podcast now

- 22:41 — AMD CEO Lisa Su’s AI projections analyzed

- 27:55 — Who will actually use AI by 2031?

- 35:45 — Why not everything needs to scale

- 39:05 — Tech CEO bubble and the global south perspective

- 41:38 — Closing remarks

Thank you for joining me on this first episode. Whether you’re curious about AI ethics, governance, or simply want to hear a thoughtful critique of where the AI industry is heading, I hope you’ll find value here.

I’m not here to dismiss artificial intelligence, it has genuinely transformative applications. Rather, I want to ensure that as AI advances, we don’t lose sight of its human and environmental costs, and that voices from communities beyond Silicon Valley get heard.

I’d love to hear your thoughts. Drop a comment, share this with someone who’d appreciate the conversation, and subscribe to stay updated as we dive deeper into these critical topics.

Subscribe now

Refer a friend

See you in the next episode.

Cheers,

Rem “Kuya Dev” Lampa

The AI Guardrail Podcast

Thank you for reading The AI Guardrail — your support allows me to keep doing this work.

If you enjoy The AI Guardrail, it would mean the world to me if you invited friends to subscribe and read with us. If you refer friends, you will receive benefits that give you special access to The AI Guardrail.

How to participate

1. Share The AI Guardrail. When you use the referral link below, or the “Share” button on any post, you'll get credit for any new subscribers. Simply send the link in a text, email, or share it on social media with friends.

Refer a friend

2. Earn benefits. When more friends use your referral link to subscribe, you’ll receive special benefits.

• Get Podcast Shoutout for 5 referrals

• Get Membership to Discord Server for 10 referrals

• Get 1-on-1 Virtual Coffee Chat for 25 referrals

Visit the leaderboard

To learn more, check out Substack’s FAQ.

Thank you for helping get the word out about The AI Guardrail!

Not yet subscribed?

A few years ago, I gave a TEDx talk in Davao City about Colonialism in the Age of AI and Big Data. Just as historical empires once grabbed land and resources, Big Tech now “explores, expands, exploits, and exterminates” using our digital lives as the raw material. We’ve moved from walled cities like Intramuros to “Data Colonies”, the smartphones in our pockets that extract our data to build wealth for an elite few, often without our consent. With the emergence of Generative AI and LLMs, the dangers of tech without guardrails brings even greater risks than ever, at a pace we’ve never seen before.

Why “The AI Guardrail Podcast” Matters Now

The AI Guardrail Podcast is part of my efforts to respond to the need for better safeguards around AI innovation and development. I’m also currently pursuing an MSc in Responsible AI and considering taking up AI Law because I believe we need a new generation of professionals who can act as bridges, able to speak both the language of code and the language of policy.

The AI Guardrail Podcast is a personal journal for:

• Engineers who want to rediscover the joy of building solutions through software by restoring tech industry’s conscience and morality.

• Business Leaders who want to navigate the evolving AI innovation and regulation landscape and their respective business risks.

• Policy Geeks who want to understand the technical reality behind AI governance frameworks like the EU AI Act, NIST, and OECD.

If you’re curious about my TEDx talk that started this journey, where I discuss “digital sweatshops,” algorithmic bias, and why AI queries consume as much water as a bottle of Evian, you can watch the talk below.

No One Has All the Answers

The AI ethics and governance space is still very much a young and emerging discipline. Like everyone else, I’m still learning and trying to make sense of where things are heading, and where things should be headed. Together, let’s try to find a way toward a more equitable digital world in the age of AI.

I’m glad you’re here for the journey. Let’s build some better guardrails.

— Rem Lampa (Kuya Dev)