Hey {{first_name|Conscious Church Fam}}
Week 2 back in the saddle. Loved the feedback last week on a. being back and b. the content of the musings.
This week: OpenAI dropped what they're calling their "best model ever" (again), the Pentagon got into a full-blown standoff with Anthropic over AI weapons, and the Supreme Court just settled whether a robot can claim copyright (spoiler: nope).
In today's recap:
GPT-5.4: OpenAI's latest "best ever" model (the 5th this year)
Anthropic vs. Pentagon: The AI ethics fight that ended with a blacklist
Supreme Court ruling: AI can't own what it creates (but that creates weird problems)
Let's dive in 👇
💭 Josh’s Musings
Building for the Kingdom vs. building for comfort
Last week I wrote about the rise of the generalist. How the messy, winding career path isn't a bug anymore. It's a feature.
But I've been chewing on something related since then.
I think a lot of us say we want meaningful work, meaningful lives, lives that build God's kingdom here on earth. But we're not actually living in ways that produce that.
We chase comfort. We chase approval. We scroll instead of ship/deliver. We optimise for what looks impressive instead of what actually serves people and ultimately glorifies God.
And then we wonder why we feel restless. Underdeveloped. Like there's more in us that we're not tapping into. Like we're capable of more than we're currently doing.
I've felt this. The pull toward the path that's easier to explain. The one that gets nods of approval but doesn't actually stretch me. The work that perhaps pays well but doesn't require everything God's put in me.
I saw a friend expand on this phrase (that I mentioned in the email last week) earlier:
"jack of all trades, master of none"
It used to sting. Like I'd failed by not honing in, refining and really developing over years.
But apparently. That's not the full quote.
The original line is: "A jack of all trades is a master of none, but oftentimes better than a master of one."
That hits different.
Because maybe polymaths aren't scattered. Maybe they're exactly what the Kingdom needs right now. People who can synthesise across domains. Who bring unusual combinations of gifts to the table. Who've practised enough different things that they can see connections others miss and use them to serve people better.
God doesn't waste seasons. He doesn't waste skills. Every bit of experience you picked up along the way, even the weird stuff that felt off brand at the time, it's all part of the toolkit you need to build what He's calling you to build.
Meaning comes from what you repeatedly do and create in service of something bigger than yourself.
Not from what you consume. Not from what you optimise for your own comfort. From what you build. What you ship. The work that requires you to show up fully and use your abilities to serve others and glorify God.
Design your life around Kingdom development instead of personal comfort. Practice things that build character. Commit to work that requires your full ability. Use your gifts in ways that serve people and point them to Jesus.
Over time, that's what makes you someone stronger, wiser, more capable. Someone who's actually building the Kingdom instead of just talking about it.
It's messy. It's uncomfortable. But it feels like obedience.
And that's where meaning lives.
And it’s where I want to be.
FYI this was a message to myself 🙂
🙌 Stay Curious, Stay Conscious, Stay Wild
Josh
LATEST NEWS
Image Source: Google Nano Banana 2
Recap:
OpenAI launched GPT-5.4 this week. Their most capable model yet. It combines elite coding (matching GPT-5.3 Codex) with the knowledge work strength of GPT-5.2, adds native computer use capabilities, and supports up to 1 million tokens of context. Available now in ChatGPT as "GPT-5.4 Thinking" and via API.
The Details:
Beats competitive models on spreadsheet modelling (87.3% vs 68.4% for GPT-5.2)
Human raters preferred its presentations 68% of the time
Half the token usage on complex tasks (offsetting higher per token price)
5th major OpenAI model update in 8 months
Early testers calling it "the best model we've ever tried"
Conscious Take:
Every few weeks, a new "best ever" ships. And the gap between those paying attention and those who aren't? It's widening fast.
Honestly at this current point I would just say find one that works best for you and your budget. There are nuances to all the models. Claude I like for writing and it’s defintely less of a YES man compared with ChatGPT (even with my custom instructions).
But Codex (OpenAI) is top for Coding currently.
So whether you’re using, Gemini, Claude, ChatGPT etc don’t worry that you’re ‘missing out’.
Image Source: Google Nano Banana 2
Recap:
Anthropic refused to remove usage restrictions from Claude for military use, triggering a standoff that ended with Trump's Pentagon declaring the AI company a "supply chain risk". A designation usually reserved for foreign adversaries like Huawei or Chinese telecoms.
The Details:
Anthropic's two non negotiable guardrails: no mass domestic surveillance, no fully autonomous weapons
Defence Secretary Pete Hegseth demanded "all lawful purposes" access with a Friday 5:01pm deadline
When Anthropic refused, Trump ordered all federal agencies to stop using Claude
The military continues using Claude during active Iran operations (because it's the best tool they have)
Public response: Claude app downloads surged, subscriptions spiked, tech workers across the industry sided with Anthropic
Anthropic vowed to challenge the designation in court
Conscious Take:
This isn't a tech story. This is a theology story.
Who gets to decide what a creation is for?
Anthropic says: "We built this, and we won't let it be used for mass surveillance or killing without human oversight."
The Pentagon says: "If it's lawful, it's ours to use. You don't get a veto."
Both are claiming authority. One rooted in conscience, one in power.
We make things in the image of the Maker. But we don't always get to control what happens to them once they're released into the world.
The difference is whether we're willing to suffer for the line we draw.
Anthropic just found out what it costs to say no.
And honestly? I respect it. Even if it costs them everything.
Image Source: Google Nano Banana 2
Recap:
The U.S. Supreme Court refused to hear a case challenging whether AI generated art can be copyrighted, ending Stephen Thaler's decade long legal battle to secure copyright for artwork created by his AI system, DABUS.
The Details:
Copyright Office, lower courts, and now SCOTUS all affirmed: copyright requires human authorship
Thaler argued corporations hold copyrights, so why not AI?
Courts disagreed: human authorship is a "bedrock requirement"
Unintended consequence: rapidly growing public domain of AI content that can't be copyrighted
This could reshape music streaming, stock imagery, and creative industries where cheaper AI generated content proliferates
Conscious Take:
The ruling protects something important: creativity belongs to persons, not processes.
But it also exposes a question the court didn't answer: What happens when humans collaborate with AI?
If I write a prompt, edit the output, integrate it into my work... am I the author?
The Copyright Office says maybe, depending on "the extent to which the human had creative control."
That's not a legal standard. That's a judgement call.
And in a world where many people are using AI to draft emails, design graphics, edit sermons? We're all making that judgement call every week.
The line between tool and co creator is thinner than we think.
"Unless the LORD builds the house, the builders labour in vain." — Psalm 127:1
Build with Kingdom purpose, not personal comfort. That's where the meaning is.
That's all for now
Stay conscious,
Josh
P.S. If you liked this then please forward it on to someone you think would enjoy it. And if someone forwarded you this and you liked it, you can sign up here.
