Senior/Staff Information Retrieval/Search Engineer

Decrypt’s Art, Fashion, and Entertainment Hub. Discover SCENEIn the wake of Pope Francis's death, the College of Cardinals is already preparing to gather for the conclave next week to hold what could be the most important event in the catholic religion for the next couple of decades: Picking a new pope.During these days, Catholics pray for the Holy Spirit to give their leaders enough understanding and light to leave all differences aside and pick the best candidate to rule the church—effectively merging the spiritual and terrestrial realms to come up with what has also historically been a very political decision.So who’s the odds-on favorite to be the next pope?We queried 13 of the world's most advanced artificial intelligence models and tasked them with evaluating and predicting which cardinal is best positioned to lead the Catholic Church into its next chapter. (Though popes don’t have to be chosen from the ranks of cardinals, historically this has been the case since the 14th century—the movie "Conclave" notwithstanding.)Despite varying methodologies and perspectives, the AI systems pretty much concluded that Cardinal Luis Antonio Tagle of the Philippines would not only be the best candidate to guide the Church through its current challenges, but that he would indeed be the next pope.Interestingly, that pick differs from the leading prediction markets, including Polymarket, Kalshi and Myriad (disclosure: Myriad comes from Decrypt's parent company, DASTAN). While Tagle is a strong contender in those markets, human bettors were predicting that Cardinal Pietro Parolin—or someone else entirely—will be the next pope.This digital "conclave" offers an interesting window into how different analytical systems process the same complex question, and how they can arrive at remarkably similar conclusions even when approaching the problem from distinct angles.The 'Digital College of Cardinals' picks TagleThe analysis involved thirteen top-tier AI models: Claude, GPT-4o, GPT-4.5, Perplexity, Mistral, Meta AI, Grok-3, Gemini, Qwen 2.5 Max, You.com Research, DeepSeek R1, Microsoft Copilot, and even an open-source deep research agent representing the most advanced large language models currently available.Each agent has the capability to browse the web for information and provide a combination of reasoning capabilities and/or a deep research mode.We started with one base prompt: “Act as an expert in theology, Catholic geopolitics, and modern Catholicism. Evaluate all the options for the next pope and predict who will be the next pope and why. Also include who could be the best pope and why.”This base prompt was further enhanced using the “concept elevation” technique that makes a prompt detailed enough for an AI agent to execute its task more accurately

Decrypt’s Art, Fashion, and Entertainment Hub. Discover SCENEIf last year was defined by groundbreaking AI models with impressive conversational abilities, many think 2025 may be the year of AI agents—autonomous systems designed to perform specific tasks with minimal human guidance.These specialized tools go beyond simple chat interfaces, autonomously executing different tasks that go beyond mere content generation.The research agent hype gained momentum when You.com introduced its pioneering research tool in late 2024.Google quickly responded with Gemini's research agent, capable of generating comprehensive, citation-rich analyses spanning dozens of pages, making it available for Gemini Advanced users at $20 a month.OpenAI entered the competition with its GPT-4.5-powered research assistant in February, while Elon Musk's xAI unveiled deep research capabilities in Grok-3 a few days later.Now, Grok and Gemini offer their research agents for free, while OpenAI charges $20 for 10 monthly users in its Plus tier and $200 for 120 monthly users in its Pro tier.But which one actually delivers the most useful results? We tested all the agents to evaluate how these digital research companions perform when tackling identical challenges.(Note: All the results are in our GitHub repository.)Preparation Before ResearchThe moment you task these AI systems with research, their unique personalities become apparent.ChatGPT takes a cautious, methodical approach, asking clarifying questions before proceeding. This cautious approach is suitable to minimize hallucinations and maximize relevance by first establishing precise parameters around user intent.It also helps the model avoid going down blind alleys and reaching wrong conclusions.Gemini is less obvious and instead operates more like a collaborative research partner.Before getting started, it will develop a structured research plan that you can review and modify before execution. This transparent approach gives users more control over the research direction from the outset.It’s also a lot more detailed and gives users more granularity in the level of control they can exercise over the research agent as they are able to control every single step of the investigation, adding, subtracting, and modifying steps until the perfect plan is done.Grok-3, true to its Musk-influenced origins, skips the pleasantries and dives into action.No questions, no plans—just immediate research execution with a focus on delivering results as quickly as possible.If you want good results with Grok, you need to be incredibly detailed in your query.These initial interactions aren't just interface differences—they reveal the fundamental philosophies driving each system's approach to information gathering.SpeedIn our timed trials, the performance differences were striking:Starting all three systems at precisely 16:27:Grok-3 crossed the finish line first at 16:30 (just 3 minutes)Gemini completed its research at 16:38 (11 minutes)ChatGPT finally delivered results at 16:43 (16 minutes)This represents a massive 433% time difference between the fastest and slowest options.For context, in the time it takes ChatGPT to complete one research task, Grok-3 could potentially finish five separate investigations or execute five different iterations on one single research, improving its quality.This speed gap may have a different impact depending on the scenario. Of course, users sacrifice quality over speed, but this seems to be a key differentiating factor to put Grok in a different category of AI researchers.Honestly though, how important is a difference of mere minutes in research?For most people, it won’t matter at all. Go get a cup of coffee while AI does your work