The Dog Coffee Fail is a 17-second AI clip of two golden retrievers in cream bibs making coffee at a kitchen counter. The big one pours, the little one spoons in the grounds and spills them, the milk misses the mug and floods the counter, and the big one covers its eyes while the little one looks up proudly, all to "Con la Misma Piedra" by Julio Iglesias. You supply both characters. This guide goes from casting to posting. When your video is back, the companion article, checking your Dog Coffee Fail result, walks through it second by second.
Put your pets behind the counter
Add two photos, review your credit quote and make the 17-second Dog Coffee Fail with the original song.
What the template gives you
A 17-second portrait clip with the original song and the reference's camera moves, including the closer view for the milk. You replace both characters. Everything else stays: the honey-wood kitchen, the glossy black counter that mirrors everything, the black coffee pot, the green mug, the milk bottle and the timing of every beat of the mess.
The ending from the reference AI reel, not an Oakgen output. The bibs in your video are left blank.
| Time (about) | What happens |
|---|---|
| 0 to 5 s | The big one pours coffee into the green mug; the little one watches |
| 5 to 10 s | The little one scoops grounds into the mug and spills them on the counter |
| 10 to 12 s | Closer view: the big one tips the milk bottle; the milk floods the counter |
| 12 to 14 s | Both look at the mess |
| 14 to 17 s | The big one covers its eyes; the little one looks up proudly |
Step 1: Cast the two roles
The big one, on the left, does the serious pouring and the facepalm. The little one, on the right, makes the mess and stays unbothered. Most people put the bigger or calmer pet in the big one's slot and the younger or cheekier one in the little one's. Two clearly different-looking subjects make it easier to spot a swap.
Step 2: Pick the photos
For each role, one subject per photo:
- Face toward the camera. Both faces are visible most of the time, and the little one looks straight up at the end.
- Front paws or hands visible. They pour, scoop, hold the mug and cover the eyes.
- Even light, no filters. Shadows on the face make the two harder to tell apart.
Both photos are required: the template does not keep either reference dog.
Step 3: Read the quote
Open the Dog Coffee Fail template, add a photo to the Big one and Little one slots and pick a quality: 480p, 720p or 1080p. The studio shows the credit price before you generate. It covers the full video, both photo preparations and three scene references. A 480p run is the cheaper way to check new photos first.
Step 4: Check the result
Watch it once with sound and once muted. Look for the two characters never swapping faces, one pool of milk on the counter, the big one's paws over its eyes at the end, and blank bibs. The second-by-second checklist covers each stretch and the one change to try if something is off.
Step 5: Post it
Set up who is to blame and let the ending land: "Me watching my roommate make coffee" or "Older sibling vs younger sibling". Keep text in the top band over the cabinets and use the calm pour as the cover, not the facepalm. More in Dog covers its eyes: coffee fail captions.
Say that the video is AI-made, and use photos only of your own pets or of people who agreed to appear.
Related reading
- The song in the Dog Coffee Fail: "Con la Misma Piedra" and what the title means.
- A cat and a dog, or you and your pet: mixed pairs and people.
- A one-photo pet trend: how to make the Wizard Cat Dance AI video.
- All the pet trends side by side: AI pet video trends compared.
Frequently asked questions
How many photos do I need for the Dog Coffee Fail?
Two: one for the big one on the left and one for the little one on the right. Both are required.
Do both have to be dogs?
No. Each role keeps the species in its photo, so a cat and a dog, two cats, or you and your pet all work.
What song is in the Dog Coffee Fail?
"Con la Misma Piedra" by Julio Iglesias. Your video keeps that original soundtrack.
What shape and length is the video?
About 17 seconds, portrait 9:16.
Is the preview video made by Oakgen?
No. It is the reference AI reel the trend comes from, not an Oakgen result.
