💡How to Prompt GPT-5 Successfully
AI Is Bringing Back the In-Person Interviews, Exams, and real human interactions.
Midweek update, friends, and the tech world’s buzzing. Musk is taking legal action against Apple, while Neuralink and SpaceX face fresh competition. Perplexity wants to buy Google’s Chrome browser (which, by the way, isn’t even for sale). And in a twist, AI is bringing something positive: a return to real human interaction.
All that, plus a quick guide on how to prompt GPT-5 like a pro. Let’s dive in, and stay curious.
How to Prompt GPT-5 Successfully
Why Perplexity Will Not Buy Chrome?
AI Tools
AI Is Bringing Back the In-Person Interviews, Exams, and real human interactions.
📰 AI News and Trends
Musk says xAI to take legal action against Apple over App Store rankings
You can now choose between “Auto”, “Fast”, and “Thinking” for GPT-5. Most users will want Auto, but the additional control will be useful for some people.
Let the poaching wars begin. Microsoft is trying to poach Meta AI talent and is offering multimillion-dollar pay packages
Anthropic is offering its Claude AI models to all three branches of the U.S. government for $1 per year, expanding on OpenAI’s earlier $1 offer to only the executive branch.
🌐 Other Tech news
CoreWeave’s revenue more than tripled to $1.2 billion in the three months
Sam Altman challenges Elon Musk with plans for Neuralink rival
SpaceX rival AST SpaceMobile prepares to deploy nearly five dozen satellites
Inside the Multimillion-Dollar Gray Market for Video Game Cheats
Scientists leave the US for China as Trump cuts funding
How to Prompt GPT-5 Successfully
GPT-5 is powerful, and it has come with some caveats. OpenAI released it and then pulled it back after users complained. Now, GPT5 is back in full force, but it only gives great results if you give it great instructions. Here’s how to craft prompts that deliver clear, accurate, and actionable answers. Some of these instructions are known to many as we have been using them on our prompts, but it is a worthwhile reminder of how to craft a successful prompt to get the results we need and save time when using a power tool.
1. Be Specific and Contextual
Instead of: "Summarize this"
Use: "Summarize this news article in under 100 words for a tech-savvy audience, highlighting numbers, financial impact, and strategic implications."
Define length, tone, audience, and focus areas.
2. Give It a Role
Telling GPT-5 who it is shapes the output.
Example: "Act as an investigative tech journalist. Write a clear, fact-based analysis of Perplexity AI’s $34.5B Chrome bid."
3. Break Complex Tasks Into Steps
Instead of asking for everything at once, guide GPT-5 through smaller steps:
Research the facts.
Create a concise summary.
Suggest three headline options.
4. Use Constraints to Get Precision
Constraints force GPT-5 to prioritize what matters.
Example: "Give me 3 bullet points with numbers or percentages, written at an 8th-grade reading level."
5. Iterate and Refine
Your first prompt is rarely perfect. Try:
“Make it more concise.”
“Add 2 more examples.”
“Rewrite for a professional audience.”
6. Feed It Reference Material
The more relevant info you give, the more accurate it gets. Copy-paste text or link to a source (if the model has browsing).
7. Test Multiple Angles
If you need creative ideas, ask for variations:
"Give me 5 alternative headlines, from serious to witty."
🧰 AI Tools
Download our list of 1000+ Tools for free.
Why Perplexity Will Not Buy Chrome?
Perplexity AI has made an unsolicited $34.5 billion all-cash bid to buy Google’s Chrome browser, despite being valued at just $18 billion. The three-year-old startup, led by Aravind Srinivas, claims multiple funds have offered to finance the deal, which makes them a more serious player in AI-driven browsing, potentially attracting more investors, talent, and partnerships.
Chrome, with over 3 billion users, is one of the most valuable gateways for search traffic and user data — both critical assets in the AI search race. Perplexity says it would keep Chromium open source, invest $3 billion over two years, and retain Google as Chrome’s default search engine. Acquiring Chrome would supercharge its AI browser, Comet, and put it head-to-head with rivals like OpenAI, which is also developing its own AI browser.
This is more of a savvy publicity play than anything else. The bid keeps Perplexity in headlines, signals to regulators that it’s a ready buyer if Google is ever forced to divest Chrome, and positions the startup as a serious contender in the browser-AI market. The DOJ has already floated Chrome’s divestiture as a potential remedy in Google’s antitrust case.
Still, Google has shown no interest in selling and plans to fight the U.S. ruling that found it held an unlawful search monopoly. Even if regulators push for divestiture, appeals could stretch the process for years. In reality, Perplexity may be better served by improving Comet and investing heavily in marketing rather than waiting for a forced sale that may never come. After all, if pushed, Google might rather shut Chrome down than hand it over to a direct competitor.
AI Is Bringing Back the In-Person Interviews, Exams, and real human interactions.
The AI boom isn’t just changing how we work; it’s reversing years of remote trends. Companies like Google, Cisco, and McKinsey are reviving face-to-face job interviews to counter a surge in AI-powered cheating.
Recruiters report candidates using off-camera AI tools to answer technical questions, and even deepfake video and voice to impersonate others. FBI investigations have revealed thousands of fake profiles, including a North Korean scheme to secure remote U.S. tech jobs.
In-person interview requests jumped from 5% in 2024 to 30% in 2025 among some recruiters.
6% of job seekers admit to interview fraud, and Gartner predicts that by 2028, 1 in 4 candidate profiles will be fake.
Some companies now use biometric verification and deepfake detection before making offers.
Google CEO Sundar Pichai says in-person rounds are now used to ensure candidates truly have the skills, especially in coding roles, where AI can instantly provide answers.
What This Means Beyond Hiring
The same AI-assisted cheating problem is creeping into schools, universities, and professional certifications. We could see:
Oral exams replace take-home essays to ensure students understand material in their own words.
Live practical tests for skills like coding, design, or engineering — graded in real time without AI assistance.
Face-to-face thesis defenses are becoming mandatory in more degree programs.
On-site assessments for professional licenses in law, medicine, and engineering to confirm human competence.
Example: If 25% of job candidates may be fake by 2028, how long before schools assume a similar percentage of AI-assisted assignments? Some universities are already piloting AI-free exam rooms with locked-down devices, while law schools in Australia have returned to 100% oral assessments for certain modules.
AI is pushing us back to real human interaction in hiring, education, and credentialing. Face-to-face assessments may become the gold standard for proving you’re truly qualified, whether for a diploma, a dream job, or a professional license.
The AI bringing back in-person interviews is interesting. I started out feeling good when I saw that, but that went away pretty quick. That's not going to help anyone who lives some distance from the place they have to interview at. That could be another city or state. I think it's great to try to address all the faking and cheating, but I also think these companies and others need to think harder and still allow video interviews.